A Windows XP help forum. PCbanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » PCbanter forum » Windows 10 » Windows 10 Help Forum
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)



 
 
Thread Tools Rate Thread Display Modes
  #31  
Old November 14th 18, 11:29 PM posted to alt.comp.os.windows-10
VanguardLH[_2_]
external usenet poster
 
Posts: 10,881
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)

NY wrote:

I'll boot the PC into its BIOS tomorrow and make a note of any version
number. I'm almost certain that it is branded "Dell".

The User's Guide says (page 79): "Virtualization (Off default) Specifies
whether a virtual machine monitor (VMM) can utilize the additional hardware
capabilities provided by Intel Virtualization technology."

The manual is dated August 2007.


Looks like you PC has a Xeon processor. I didn't look at all of them to
check if all models and versions of the Xeon processor support the VT-x
virtualization mode. I looked only at the latest model and it does.

https://www.intel.com/content/www/us...rocessors.html

Supposedly that leads you to information on whether or not your
processor supports virtualization. Just enabling an option in BIOS
won't make a feature appear in the processor that it doesn't have.
Could be old versions of Xeons did not support VT-x and why the option
was disabled in the BIOS, by default.

https://www.intel.com/content/www/us...r-numbers.html

That shows how to interpret the processor's number to a family model
line; however, the only entry on that Microsoft page is for the mobile
versions of the Xeon, and you have a desktop.

You could use Piriform's Speccy to get info on the CPU. It doesn't list
the model number but gives the processor title and its code name which
might focus in on finding the specs for your particular CPU. For my
CPU, it reports "Virtualization: Supported, Disabled". I have a
salvaged Acer PC with almost no user configuration in its BIOS, so that
box was designed to not support VT-x no matter what CPU was put in the
mobo's CPU socket. I look again later but remember checking before in
the BIOS if there was a virtualization setting that I could enable but
didn't find one.

I also have the CPU-z tool. It says my CPU (Intel Core 2 Quad Q9400
Yorkfield) does support VT-x instructions. However, since there is no
BIOS support for it, I can't use it. This is an example of where I
noted that the BIOS may not yet support all the functionality of the
hardware. A later BIOS update may add that support. For me, there
haven't been any BIOS updates since 2009 for my PC and I'm on the latest
BIOS version that Acer ever released for this PC.

Since your BIOS does have a VT-x setting, next is to determine if your
processor supports it. Instead of delving into Microsoft online
documents trying to dig up data that says whether or not VT-x is
supported by your CPU, just use Speccy or CPU-z or both. Support must
be present in *both* the BIOS and the CPU. Also, both Speccy and CPU-z
will show motherboard/mainboard information, like the BIOS brand and
version (in case the POST screen disappears so fast that you cannot read
the line showing BIOS brand and version).

I use Avast free. Under its Troubleshooting settings is "Use hardware-
assisted virtualization" (and a sub-option "Use nested virtualization
where possible"). Presumably this is to help with isolating Avast's NG
component (deepscreen, sandbox, and safezone); however, the feature is
minimal in the freeware version and it's in the payware version where
you really get a sandbox for unknown processes to get tested, so I don't
know why the option appears in the freeware version.

https://malwaretips.com/threads/avas...40/post-281750

While my CPU supports virtualization instructions, my BIOS does not, so
this option is neutered in Avast. There have been bugs with this
feature in Avast where users (where both BIOS and CPU support
virtualization) reported that HAV (hardware-assisted virtualization) was
disabled when Avast had this option enabled, so disabling this option
allowed virtual machines to work again because Avast was no longer
interferring with HAV. I probably never ran into Avast's HAV
interference: its HAV option is enabled, my CPU supports HAV, but my
BIOS doesn't so I don't get to use HAV on my home PC.
Ads
  #32  
Old November 14th 18, 11:41 PM posted to alt.comp.os.windows-10
VanguardLH[_2_]
external usenet poster
 
Posts: 10,881
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)

VanguardLH wrote:

NY wrote:

I'll boot the PC into its BIOS tomorrow and make a note of any version
number. I'm almost certain that it is branded "Dell".

The User's Guide says (page 79): "Virtualization (Off default) Specifies
whether a virtual machine monitor (VMM) can utilize the additional hardware
capabilities provided by Intel Virtualization technology."

The manual is dated August 2007.


Looks like you PC has a Xeon processor. I didn't look at all of them to
check if all models and versions of the Xeon processor support the VT-x
virtualization mode. I looked only at the latest model and it does.

https://www.intel.com/content/www/us...rocessors.html

Supposedly that leads you to information on whether or not your
processor supports virtualization. Just enabling an option in BIOS
won't make a feature appear in the processor that it doesn't have.
Could be old versions of Xeons did not support VT-x and why the option
was disabled in the BIOS, by default.

https://www.intel.com/content/www/us...r-numbers.html

That shows how to interpret the processor's number to a family model
line; however, the only entry on that Microsoft page is for the mobile
versions of the Xeon, and you have a desktop.

You could use Piriform's Speccy to get info on the CPU. It doesn't list
the model number but gives the processor title and its code name which
might focus in on finding the specs for your particular CPU. For my
CPU, it reports "Virtualization: Supported, Disabled". I have a
salvaged Acer PC with almost no user configuration in its BIOS, so that
box was designed to not support VT-x no matter what CPU was put in the
mobo's CPU socket. I look again later but remember checking before in
the BIOS if there was a virtualization setting that I could enable but
didn't find one.

I also have the CPU-z tool. It says my CPU (Intel Core 2 Quad Q9400
Yorkfield) does support VT-x instructions. However, since there is no
BIOS support for it, I can't use it. This is an example of where I
noted that the BIOS may not yet support all the functionality of the
hardware. A later BIOS update may add that support. For me, there
haven't been any BIOS updates since 2009 for my PC and I'm on the latest
BIOS version that Acer ever released for this PC.

Since your BIOS does have a VT-x setting, next is to determine if your
processor supports it. Instead of delving into Microsoft online
documents trying to dig up data that says whether or not VT-x is
supported by your CPU, just use Speccy or CPU-z or both. Support must
be present in *both* the BIOS and the CPU. Also, both Speccy and CPU-z
will show motherboard/mainboard information, like the BIOS brand and
version (in case the POST screen disappears so fast that you cannot read
the line showing BIOS brand and version).

I use Avast free. Under its Troubleshooting settings is "Use hardware-
assisted virtualization" (and a sub-option "Use nested virtualization
where possible"). Presumably this is to help with isolating Avast's NG
component (deepscreen, sandbox, and safezone); however, the feature is
minimal in the freeware version and it's in the payware version where
you really get a sandbox for unknown processes to get tested, so I don't
know why the option appears in the freeware version.

https://malwaretips.com/threads/avas...40/post-281750

While my CPU supports virtualization instructions, my BIOS does not, so
this option is neutered in Avast. There have been bugs with this
feature in Avast where users (where both BIOS and CPU support
virtualization) reported that HAV (hardware-assisted virtualization) was
disabled when Avast had this option enabled, so disabling this option
allowed virtual machines to work again because Avast was no longer
interferring with HAV. I probably never ran into Avast's HAV
interference: its HAV option is enabled, my CPU supports HAV, but my
BIOS doesn't so I don't get to use HAV on my home PC.


Oh, it's not just virtuals machines (VMWare Player, VirtualBox) where
compounding the virtualization layers by enabling the one in Avast
causes problems. See:

https://support.bluestacks.com/hc/en...acks-run-slow-

Anything that uses HAV can get screwed by Avast's NG. If you're using
VMs to test unknown or untrusted software, you don't need Avast's HAV
support. Nesting is supposed to help but users still report problems
with Avast's NG component. Although Avast's HAV option is enabled (by
default) in my setup of it, and despite that HAV is not available on my
PC because the BIOS doesn't support it, I'll probably disable Avast's
HAV in case I migrate my setup to a new box where the BIOS does support
HAV.

There should only be one hypervisor running on the OS. Avast should
stop using their own and implement the one included in Windows 10,
similar to how backup programs use VSS supplied by Microsoft and
included in Windows rather than roll their own solution. Of course, to
be ubiquitous across all editions of Windows 10, Microsoft would need to
add Hyper-V to the Home edition.

https://docs.microsoft.com/en-us/vir...-v-on-windows/
https://en.wikipedia.org/wiki/Hypervisor
  #33  
Old November 15th 18, 10:31 AM posted to alt.comp.os.windows-10
NY
external usenet poster
 
Posts: 586
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)

"VanguardLH" wrote in message
...
NY wrote:

I'll boot the PC into its BIOS tomorrow and make a note of any version
number. I'm almost certain that it is branded "Dell".

The User's Guide says (page 79): "Virtualization (Off default)
Specifies
whether a virtual machine monitor (VMM) can utilize the additional
hardware
capabilities provided by Intel Virtualization technology."

The manual is dated August 2007.


You could use Piriform's Speccy to get info on the CPU. It doesn't list
the model number but gives the processor title and its code name which
might focus in on finding the specs for your particular CPU.


The summary page for Speccy says:

Operating System
Windows 10 Home 64-bit
CPU
Intel Xeon E5420 @ 2.50GHz 48 °C
Harpertown 45nm Technology
RAM
36.0GB FB-DDR2 @ 332MHz (5-5-5-15)
Motherboard
Dell Inc. 0RW199 (CPU)
Graphics
DELL 2707WFP (1920x1200@59Hz)
512MB NVIDIA Quadro FX 1700 (NVIDIA) 50 °C
Storage
298GB ATA SAMSUNG HD321KJ SCSI Disk Device (SATA ) 25 °C
298GB ATA SAMSUNG HD321KJ SCSI Disk Device (SATA ) 23 °C
698GB ATA Hitachi HDS72107 SCSI Disk Device (SATA ) 30 °C
Optical Drives
HL-DT-ST DVD+-RW GSA-H73N
TSSTcorp DVD-ROM TS-H353B
Audio
High Definition Audio Device

The CPU/motherboard/RAM sections say

CPU
Intel Xeon E5420
Cores 4
Threads 4
Name Intel Xeon E5420
Code Name Harpertown
Package Socket 771 LGA
Technology 45nm
Specification Intel Xeon CPU E5420 @ 2.50GHz
Family 6
Extended Family 6
Model 7
Extended Model 17
Stepping 6
Revision C0
Instructions MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, Intel 64, NX, VMX
Virtualization Supported, Enabled
Hyperthreading Not supported
Rated Bus Speed 1330.0 MHz
Stock Core Speed 2500 MHz
Stock Bus Speed 333 MHz
Average Temperature 48 °C



RAM
Memory slots
Total memory slots 16
Used memory slots 8
Free memory slots 8
Memory
Type FB-DDR2
Size 36864 MBytes
DRAM Frequency 332.5 MHz
CAS# Latency (CL) 5 clocks
RAS# to CAS# Delay (tRCD) 5 clocks
RAS# Precharge (tRP) 5 clocks
Cycle Time (tRAS) 15 clocks
Bank Cycle Time (tRC) 20 clocks
Physical Memory
Memory Usage 6 %
Total Physical 36 GB
Available Physical 34 GB
Total Virtual 71 GB
Available Virtual 69 GB
SPD
Number Of SPD Modules 8
Slot #1
Type FB-DDR2
Size 8192 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 1131222199
Week/year 08 / 09
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #2
Type FB-DDR2
Size 1024 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 672734829
Week/year 07 / 08
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #3
Type FB-DDR2
Size 8192 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 1131230371
Week/year 08 / 09
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #4
Type FB-DDR2
Size 1024 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 672734982
Week/year 07 / 08
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #5
Type FB-DDR2
Size 8192 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 1165052533
Week/year 26 / 09
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #6
Type FB-DDR2
Size 1024 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 672734971
Week/year 07 / 08
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #7
Type FB-DDR2
Size 8192 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 1165050791
Week/year 26 / 09
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Slot #8
Type FB-DDR2
Size 1024 MBytes
Manufacturer Samsung
Max Bandwidth PC2-5300 (333 MHz)
Serial Number 672734983
Week/year 07 / 08
Timing table
JEDEC #1
Frequency 333.3 MHz
CAS# Latency 5.0
RAS# To CAS# 5
RAS# Precharge 5
tRAS 15
tRC 20
Motherboard
Manufacturer Dell Inc.
Model 0RW199 (CPU)
Chipset Vendor Intel
Chipset Model 5400B
Chipset Revision C0
Southbridge Vendor Intel
Southbridge Model 6321ESB
Southbridge Revision 09
BIOS
Brand Dell Inc.
Version A01
Date 31/01/2008
PCI Data
Slot PCI-E
Slot Type PCI-E
Slot Usage Available
Data lanes x4
Slot Designation SLOT1
Characteristics 3.3V, PME
Slot Number 0
Slot PCI-E
Slot Type PCI-E
Slot Usage In Use
Data lanes x16
Slot Designation SLOT2
Characteristics 3.3V, PME
Slot Number 1
Slot PCI
Slot Type PCI
Slot Usage Available
Bus Width 32 bit
Slot Designation SLOT3
Characteristics 5V, 3.3V, PME
Slot Number 2
Slot PCI-E
Slot Type PCI-E
Slot Usage Available
Data lanes x16
Slot Designation SLOT4
Characteristics 3.3V, PME
Slot Number 3
Slot PCI-X
Slot Type PCI-X
Slot Usage Available
Bus Width 64 bit
Slot Designation SLOT5
Characteristics 3.3V, PME
Slot Number 4
Slot PCI-X
Slot Type PCI-X
Slot Usage Available
Bus Width 64 bit
Slot Designation SLOT6
Characteristics 3.3V, PME
Slot Number 5
Slot PCI-X
Slot Type PCI-X
Slot Usage Available
Bus Width 64 bit
Slot Designation SLOT7
Characteristics 3.3V, PME
Slot Number 6


So virtualisation is supported and enabled.


CPU-Z says:

Socket 1 ID = 0
Number of cores 4 (max 4)
Number of threads 4 (max 4)
Name Intel Xeon E5420
Codename Harpertown
Specification Intel(R) Xeon(R) CPU E5420 @ 2.50GHz
Package (platform ID) Socket 771 LGA (0x6)
CPUID 6.7.6
Extended CPUID 6.17
Core Stepping C0
Technology 45 nm
Core Speed 2493.5 MHz
Multiplier x Bus Speed 7.5 x 332.5 MHz
Base frequency (cores) 332.5 MHz
Base frequency (ext.) 332.5 MHz
Rated Bus speed 1329.9 MHz
Stock frequency 2500 MHz
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, EM64T, VT-x
Microcode Revision 0x60F
L1 Data cache 4 x 32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache 4 x 32 KBytes, 8-way set associative, 64-byte line
size
L2 cache 2 x 6144 KBytes, 24-way set associative, 64-byte line size
Max CPUID level 0000000Ah
Max CPUID ext. level 80000008h
Cache descriptor Level 1, D, 32 KB, 1 thread(s)
Cache descriptor Level 1, I, 32 KB, 1 thread(s)
Cache descriptor Level 2, U, 6 MB, 2 thread(s)
FID/VID Control yes
FID range 6.0x - 7.5x
Max VID 1.200 V


IBRS not supported
IBPB not supported
STIBP not supported
RDCL_NO no
IBRS_ALL not supported

Temperature 0 51 degC (123 degF) (Core #0)
Temperature 1 47 degC (116 degF) (Core #1)
Temperature 2 45 degC (113 degF) (Core #2)
Temperature 3 53 degC (127 degF) (Core #3)
Clock Speed 0 2493.51 MHz (Core #0)
Clock Speed 1 2493.51 MHz (Core #1)
Clock Speed 2 2493.51 MHz (Core #2)
Clock Speed 3 2493.51 MHz (Core #3)
Core 0 max ratio (effective) 7.5
Core 1 max ratio (effective) 7.5
Core 2 max ratio (effective) 7.5
Core 3 max ratio (effective) 7.5


and

BIOS
-------------------------------------------------------------------------

UEFI No

Chipset
-------------------------------------------------------------------------

Northbridge Intel 5400B rev. C0
Southbridge Intel 6321ESB rev. 09
Graphic Interface PCI-Express
PCI-E Link Width x16
PCI-E Max Link Width x16
Memory Type FB-DDR2
Memory Size 36 GBytes
Memory Frequency 332.5 MHz (1:1)
CAS# latency (CL) 5.0
RAS# to CAS# delay (tRCD) 5
RAS# Precharge (tRP) 5
Cycle Time (tRAS) 15
Bank Cycle Time (tRC) 20




DMI BIOS vendor Dell
Inc.
version A01
date 01/31/2008
ROM size 1024 KB

DMI System Information manufacturer Dell
Inc.
product Precision WorkStation T7400
version unknown
serial xxxx
UUID xxxxxx
SKU unknown
family unknown

DMI Baseboard vendor Dell
Inc.
model 0RW199
revision unknown
serial xxxxxx



So Dell BIOS version A01 31 Jan 2008.

  #34  
Old November 15th 18, 11:46 AM posted to alt.comp.os.windows-10
VanguardLH[_2_]
external usenet poster
 
Posts: 10,881
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)

NY wrote:

CPU
Intel Xeon E5420
Virtualization Supported, Enabled

So virtualisation is supported and enabled.


Yep, enabled in both the BIOS (after you changed it from disabled) and
supported by the CPU. I still don't know why that would affect the
"hardware reserved" value on system RAM available.

CPU-Z says:

Socket 1 ID = 0
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, EM64T, VT-x


VT-x is the instruction set for virtualization support.

BIOS

DMI BIOS vendor Dell
Inc.
version A01
date 01/31/2008

So Dell BIOS version A01 31 Jan 2008.


Hmm, the page to which I linked said the latest BIOS version was A11.
Again, that download page at Dell that I found is:

https://www.dell.com/support/home/us...t7400/diagnose

The A11 BIOS update was the 3rd down in the blue hyperlinked list. I
didn't bother selecting an OS because: (1) The OS is irrelevant when
looking a the BIOS; and, (2) Windows 10 isn't listed as supported for
the OS-specific downloads. The hyperlink to the A11 download points to:

https://www.dell.com/support/home/us...recision-t7400
Release Date: 28 Jun 2012

That's a BIOS version over 4 years later than what you have now. Since
you got the memory all available, I wouldn't bother doing a BIOS update
now. Just consider it something you might want to do later but only if
the desktop PC is attached to a UPS. If it were me, I'd do some
research to see if Dell's BIOS updater will offer to save the current
BIOS code into a .bin file both as a backup and in case the new BIOS has
problems, so I'd have an escape route to get back to what was there
before.

To me, that enabling the virtualization option in the BIOS then released
the hardware (BIOS) reserved memory indicates something screwed up in
the old BIOS code. Dell doesn't list every BIOS version from your A01
to their latest A11 to see what changes were made in each to, for
example, see when they might've discovered a memory allocation error
with their virtualization code in BIOS - *if* that was the only setting
you changed in the BIOS. Of course, making changes in the BIOS settings
means copying the tables into the CMOS memory from where the settings
are actually used which is why I suggested replacing the CMOS battery.
If the CMOS is corrupted, there could be other BIOS settings in the CMOS
copy that don't match the settings in the EEPROM for the BIOS.
  #35  
Old November 15th 18, 11:48 AM posted to alt.comp.os.windows-10
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows(Dell Precision T7400, upgraded from Win 7 Home)

NY wrote:
"VanguardLH" wrote in message
...
NY wrote:

I'll boot the PC into its BIOS tomorrow and make a note of any version
number. I'm almost certain that it is branded "Dell".

The User's Guide says (page 79): "Virtualization (Off default) Specifies
whether a virtual machine monitor (VMM) can utilize the additional
hardware
capabilities provided by Intel Virtualization technology."

The manual is dated August 2007.


You could use Piriform's Speccy to get info on the CPU. It doesn't list
the model number but gives the processor title and its code name which
might focus in on finding the specs for your particular CPU.


The summary page for Speccy says:


DMI BIOS vendor Dell Inc.
version A01
date 01/31/2008
ROM size 1024 KB

So Dell BIOS version A01 31 Jan 2008.


https://www.dell.com/support/home/us...driverid=1dt11

Dell Precision T7400 System BIOS, A02 Release date 04 Jun 2008
Added support for 8 GB memory DIMM.

I went through the list by Googling, and that's
the only incriminating evidence I could find.
Maybe the BIOS probing sequence artificially
stopped at the 4GB mark.

You could test this "modus operandi" by briefly removing
the virtualization check, and see whether the 36GB
installed now, is cut in half to 18, or instead
reports 20GB (16+4). If it reported 20GB, that would
suggest only half of each 8GB DIMM is detected
properly with the virtualization setting changed.

Seeing 18GB would suggest some weird mirroring theory,
seeing 20GB would be suggestive of a more palatable
explanation. That A01 doesn't sense 8GB DIMMs correctly.
4+4+4+4+1+1+1+1 = 20GB.

The BIOS checks the declaration ROM (SPD) first. It
says 8GB. But the BIOS then uses the probe technique
to measure the "actual" DIMM size - this is a legacy
method used before computers had SPDs. And the legacy
method is *excellent* insurance, as very few boxes
tip over at BIOS level because of detection issues.
The result at OS might not be optimal, but by the
BIOS using the probing method, it "does the very
best that it can". It never delivers more RAM than
is actually there. You couldn't write better code
than that.

Paul
  #36  
Old November 15th 18, 01:03 PM posted to alt.comp.os.windows-10
NY
external usenet poster
 
Posts: 586
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)

"VanguardLH" wrote in message
...
The A11 BIOS update was the 3rd down in the blue hyperlinked list. I
didn't bother selecting an OS because: (1) The OS is irrelevant when
looking a the BIOS; and, (2) Windows 10 isn't listed as supported for
the OS-specific downloads. The hyperlink to the A11 download points to:

https://www.dell.com/support/home/us...recision-t7400
Release Date: 28 Jun 2012

That's a BIOS version over 4 years later than what you have now. Since
you got the memory all available, I wouldn't bother doing a BIOS update
now. Just consider it something you might want to do later but only if
the desktop PC is attached to a UPS. If it were me, I'd do some
research to see if Dell's BIOS updater will offer to save the current
BIOS code into a .bin file both as a backup and in case the new BIOS has
problems, so I'd have an escape route to get back to what was there
before.


I think now I've got the PC working to full spec, I'll leave well alone and
quit while I'm ahead. I always have a fear of "bricking" a device that needs
a BIOS/firmware update: it only takes a hiccup in the mains power to
interrupt it half-way through, so you can't go forwards or backwards.
Battery/mains devices can be powered from a battery, but that's not possible
for a desktop PC. With all the cooling fans, and the amount of heat that
needs those fans to dissipate it, a battery probably wouldn't last very
long.

Cautionary tale: when my wife bought the PC she also bought a 700 VA UPS. We
never actually tested it at the time, but a year or so later we tried it to
see how long its charge would last. The thing was totally clappped-out. The
battery would not hold charge to power a 60 W light bulb for more than a
couple of seconds. We really should have tested it when it was new, in case
it was faulty and neede to be replaced under warranty, and then kept it
charged and put it through a few discharge cycles. I believe that lead-acid
batteries don't like being either permanently on charge (which this wasn't)
or left totally discharged (which this was).

I'm not sure how long a 700 VA UPS would keep a PC running. Hopefully long
enough for it to undergo a graceful shutdown.

  #37  
Old November 15th 18, 01:44 PM posted to alt.comp.os.windows-10
VanguardLH[_2_]
external usenet poster
 
Posts: 10,881
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows (Dell Precision T7400, upgraded from Win 7 Home)

NY wrote:

"VanguardLH" wrote in message
...
The A11 BIOS update was the 3rd down in the blue hyperlinked list. I
didn't bother selecting an OS because: (1) The OS is irrelevant when
looking a the BIOS; and, (2) Windows 10 isn't listed as supported for
the OS-specific downloads. The hyperlink to the A11 download points to:

https://www.dell.com/support/home/us...recision-t7400
Release Date: 28 Jun 2012

That's a BIOS version over 4 years later than what you have now. Since
you got the memory all available, I wouldn't bother doing a BIOS update
now. Just consider it something you might want to do later but only if
the desktop PC is attached to a UPS. If it were me, I'd do some
research to see if Dell's BIOS updater will offer to save the current
BIOS code into a .bin file both as a backup and in case the new BIOS has
problems, so I'd have an escape route to get back to what was there
before.


I think now I've got the PC working to full spec, I'll leave well alone and
quit while I'm ahead. I always have a fear of "bricking" a device that needs
a BIOS/firmware update: it only takes a hiccup in the mains power to
interrupt it half-way through, so you can't go forwards or backwards.
Battery/mains devices can be powered from a battery, but that's not possible
for a desktop PC. With all the cooling fans, and the amount of heat that
needs those fans to dissipate it, a battery probably wouldn't last very
long.

Cautionary tale: when my wife bought the PC she also bought a 700 VA UPS. We
never actually tested it at the time, but a year or so later we tried it to
see how long its charge would last. The thing was totally clappped-out. The
battery would not hold charge to power a 60 W light bulb for more than a
couple of seconds. We really should have tested it when it was new, in case
it was faulty and neede to be replaced under warranty, and then kept it
charged and put it through a few discharge cycles. I believe that lead-acid
batteries don't like being either permanently on charge (which this wasn't)
or left totally discharged (which this was).

I'm not sure how long a 700 VA UPS would keep a PC running. Hopefully long
enough for it to undergo a graceful shutdown.


Even if the case is snapped together, I've been able to get apart any
UPS so that I could replace the battery. I can then replace the battery
with a new one. Although cheaper to replace the battery, the gel
batteries in the so-called sealed units end up costing nearly half the
cost of a replacement UPS. On the large units, like the 2 kVA one I
used to have, the batteries were in a drawer that could be slid out of
the case for replacement.

Most UPSes come with software that communicates from the PC to the UPS.
They will show you the estimate up-time while on battery. It may not be
super accurate but it's better than using the online calculators on
up-time for a UPS. Some software and UPS have a test mode: they
momentarily disconnect from the A/C power source and do a battery load
test. You could get a kill-o-watt meter to plug into wherever the
computer plugs into for A/C power to measure the actual drain at the
plug. I got one from a local hardware store for ~$20 to check the power
draw by a new refrigerator but I can use it on any A/C powered device.

The CMOS battery is an easy replace in a desktop PC. Because several
devices I own use the same CR-2032 coin cell alkaline battery, I bought
a pack of 20 at eBay for under $8 (but make sure the packaging shown in
the auction matches authentic manufacturer packaging and that's what you
get since there are faux batteries being sold there and elsewhere).
  #38  
Old November 15th 18, 01:46 PM posted to alt.comp.os.windows-10
Andy Burns[_6_]
external usenet poster
 
Posts: 1,318
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows(Dell Precision T7400, upgraded from Win 7 Home)

NY wrote:

I always have a fear of "bricking" a device that needs a BIOS/firmware
update: it only takes a hiccup in the mains power to interrupt it
half-way through, so you can't go forwards or backwards.


Most BIOSes have a failsafe where you can boot from a USB or FLOPPY(!)
to restore a failed flash, even without a screen or keyboard

I'm not sure if writing to the flash chips is "atomic" so that a given
block is ether written, or it is not written, and is never partly written?

Other machines (but not that DELL as far as I could see) have a dual
BIOS, so they upgrade the "half" that's not currently in use, so you
don't overwrite it, you switch-over to it afteer writing.

All that said, I do generally have machines on a UPS while flashing
firmware.
  #39  
Old November 15th 18, 02:40 PM posted to alt.comp.os.windows-10
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows(Dell Precision T7400, upgraded from Win 7 Home)

Andy Burns wrote:
NY wrote:

I always have a fear of "bricking" a device that needs a BIOS/firmware
update: it only takes a hiccup in the mains power to interrupt it
half-way through, so you can't go forwards or backwards.


Most BIOSes have a failsafe where you can boot from a USB or FLOPPY(!)
to restore a failed flash, even without a screen or keyboard

I'm not sure if writing to the flash chips is "atomic" so that a given
block is ether written, or it is not written, and is never partly written?

Other machines (but not that DELL as far as I could see) have a dual
BIOS, so they upgrade the "half" that's not currently in use, so you
don't overwrite it, you switch-over to it afteer writing.

All that said, I do generally have machines on a UPS while flashing
firmware.


The original design intent, was the "boot block" in the BIOS
was never supposed to be erased. If an attempt to erase and
program the "main BIOS block" failed, using the boot block
was supposed to allow recovery.

A downside of "boot block", is it has no graphics driver
(couldn't use VESA mode on the graphics card), so you have
to operate the controls without any text on the LCD monitor
screen. This takes practice and imagination.

There are a number of modern devices with this boot block feature.
If the usage discipline is followed, it works very well. This is
why these other devices, when people reprogram them, they
hardly ever get bricked.

However, a ton of motherboard BIOS updater designs, *insist* on
deleting the boot block first thing. Then, if the upgrade fails
for any reason... the boot block is empty, the computer is a
brick.

Sure, there might be a very good reason to change something
in the boot block, but with a little testing, you could
also throw together a boot block that would last for the
life of the motherboard.

On Gigabyte motherboards, they have a "dual BIOS". This
allows programming one ROM, while using the other ROM
for boot up. However, one thing to keep in mind - as
far as I know, the two ROMs give you a total of
one boot block and two main BIOS blocks, so there is
still a shocking lack of redundancy. If that boot
block were to fail, switching to the other ROM
wouldn't help you. We had a product at work with
dual BIOS (not really a BIOS but all the device
code is in it), which is completely redundant,
and the ROMs are equal in design. And a dedicated
arbiter selects the ROM to use. I never did any
analysis of that thing, to see what holes it had,
but the staff seemed to like it. At least that
design didn't "subset" one of the ROMs.

There was also a product called "BIOS Savior", which
was perhaps made in Taiwan. It consisted of an adapter
with a ROM in it. You then had the equivalent of the
Gigabyte setup (only the ROMs were perfectly equal and
a switch selected the one you wanted to use). I thought
this was a great invention, and overclockers used to
buy them for hacking projects. The problem there was,
the Savior was fine in the socketed PLCC era, but when
serial 8 pin EEPROMS came out, there was no BIOS Savior
design for those. It's possible the OPs system is old
enough to still have a PLCC in it. Different models
had different size EEPROMS on the adapter.

http://www.euclideanspace.com/tech/p...vior/index.htm

My newest system has *the* best feature :-) It can't
be bricked. It has a dedicated USB to serial EEPROM
chip onboard. You can flash the motherboard even when
no CPU or RAM are installed in the motherboard. You
plug in a USB stick with the BIOS image on it, push
a button, and it upgrades. One of the USB sockets has
a dotted white line around it, which indicates the
socket with the "magic" properties. This is a feature
on some Asus boards. I've not had the opportunity
to search for the chip and get a part number, and
figure out the design (the CPU heatsink is huge
and access is limited now). I was too busy proving
the motherboard was fully function when I bought it,
and didn't have time to play around with it then.

Paul
  #40  
Old November 15th 18, 08:06 PM posted to alt.comp.os.windows-10
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default Windows 10 Home PC 32 GB RAM but only 16 GB available to Windows(Dell Precision T7400, upgraded from Win 7 Home)

NY wrote:


Cautionary tale: when my wife bought the PC she also bought a 700 VA
UPS. We never actually tested it at the time, but a year or so later we
tried it to see how long its charge would last. The thing was totally
clappped-out. The battery would not hold charge to power a 60 W light
bulb for more than a couple of seconds. We really should have tested it
when it was new, in case it was faulty and neede to be replaced under
warranty, and then kept it charged and put it through a few discharge
cycles. I believe that lead-acid batteries don't like being either
permanently on charge (which this wasn't) or left totally discharged
(which this was).

I'm not sure how long a 700 VA UPS would keep a PC running. Hopefully
long enough for it to undergo a graceful shutdown.


The UPS company will have a curve which shows loading
in watts versus "minutes remaining".

The battery in mine is 12V 12Ah. That's 144Wh.
If I had your UPS, and applied a 700W load to it,
then 144Wh/700W = 0.2h = 12 minutes. The manufacturer
curve would probably say 6 minutes, just to give
some idea how much fudging is involved. When you
draw 700W from a 12V battery, that's a current flow
of 58A, which is a little high for a pint size SLA
battery. The internal resistance of the battery
will probably drop the terminal voltage below 12V.

And that's how UPSes check the battery. Mine does
an "impedance test" once a day, to check the resistance.
The test lasts for a second or two. It draws a high enough
current so the battery terminal voltage can be checked under
load. If the battery fails the test, my unit lets out
one sad little "beep" and there are no visual indicators
that latches to show the status later.

Mine was beeping once a day for a number of weeks.
At first, I thought the beep might be coming from
my carbon monoxide detector. But it turned out to
be the UPS. I bought a replacement battery online
and replaced it. It's been well behaved since.

If you need to measure the load being applied to the
UPS, you can do that with a Kill-A-Watt meter (P4400).
They're cheap enough now to be in the same ballpark
as a Harbor Freight multimeter. The Kill-A-Watt
measures the Power in units of W and VA, which allows
comparing the numbers to the W and VA ratings of
the UPS.

My newest computer idles at around 100W or so.
The older one draws only 65W or so. Once you
have some measured values, it makes it easier
to estimate what a good value for holdup time is.

Paul
 




Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off






All times are GMT +1. The time now is 12:25 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 PCbanter.
The comments are property of their posters.