Page 1 of 13 123411 ... LastLast
Results 1 to 25 of 316

Thread: p4gd1 torture chamber

  1. #1
    Xtreme X.I.P.
    Join Date
    Nov 2002
    Location
    Shipai
    Posts
    31,147
    .




    WARNING!
    for some reason memory keeps dieing on boards using the ct479, waus-mod now found out that the spd chips on the memory sticks probably just got erased/corrupted and that its possible to fix this within 5 minutes!
    heres the guide:
    http://www.xtremesystems.org/forums/...ad.php?t=85533






    welcome to the p4gd1 torture chamber

    i split this off from the p4gd1 in the us thread and wil update this first post and add all the important infos and mods and tweaks to it

    ---

    Tips, Tricks and Tweaks:

    ----.-TIP: go to advanced -> jumper free configuration -> ai overclocking and set it to manual to get all the overclocking options
    ------- TIP: enable the pci lock in this menu by setting the pci speed to 33.33mhz
    ------- TIP: disable spread spectrum
    ---- TIP: dram frequency auto = sets a 1:1 fsb:dram divider
    ------------. fsb100:dram400 = actually sets an fsb:dram 5:4 divider
    ------------. fsb100:dram533 is actually the fsb:dram 1:1 divider again.
    ------------. i tried all settings and combinations, setting the ct470 to 133fsb or 100fsb, booting at 133fsb or 100fsb or 101fsb
    ------------. and using pc2700 and pc3200 memory but i could get any other divider working.
    ------- TIP: go to advanced -> chipset configuration and set configure dram timings by spd to disabled to be able to set memory timings
    ---- TIP: 2.0-2-2 doesnt work for me with 2.9v even at only 150mhz. i had to set it to 2.0-2-3 to be 100% stable.
    ------- TIP: hyperpath 2 enabled or disabled didnt make a difference at all to me
    ------- TIP: the actualy vcore is a lot higher than what is actually set in bios
    ------------. this is with a pentium m banaias 130nm:
    ------------. 1.1125v = ~1.200v
    ------------. 1.2125v = ~1.290v
    ------------. 1.3375v = ~1.410v
    ------------. 1.4375v = ~1.504v
    ------------. 1.5375v = ~1.600v
    ------------. 1.6625v = ~1.710v
    ------- TIP: the voltage readings in bios are pretty accurate, so if you increase vcore, save to bios, reboot and enter bios again to see how
    ------------. high vcore is actually set.
    ------- TIP: vdd aka the chipset voltage is almost 0.2v lower than it should be according to the bios settings
    ------------. to get really high fsbs you need a vdd mod to around 1.8v or higher
    ------- TIP: vdimm is pretty accurate, its 0.02v higher than whats set in bios.
    ------------. to get more than 2.9v you can do the 3.3v rail to vdimm mod wich is pretty easy.
    ------- TIP: go to advanced -> onboard devices configuration and disable the ITE ide raid controller to speed up the boot process
    ------- TIP: go to advanced -> onboard devices configuration and disable all serial ports, game ports, onboard sound and onboard lan
    ------------. to make sure nothing is holding you back when ocing to the max
    ------- TIP: go to advanced -> usb configuration and disable the usb controllers to make sure its not holding you back when ocing to the max
    ------- TIP: use the EIST tool or CrystalCPUID to increase your multiplier and then increase the fsb bit by bit to find the max stable cpu speed
    ------- TIP: use the EIST tool or CrystalCPUID to lower your multiplier to find the max fsb/dram speed.
    ------------. if you think your memory is holding you back set the fsb100:dram400 divider wich results in a 5:4 fsb:dram divider to see if
    ------------. it was really the memory holding you back.
    ------- TIP: if you hit the fsb limit on your board increase the chipset voltage
    ------- TIP: if you hit the fsb limit on your board increase the pciE speed by 5mhz and see if you can get a higher fsb
    ------- TIP: it seems that a ~2,2:1 ratio for fsb:pciE speeds is needed to get the fsb stable.
    ------- TIP: if you cant increase the pciE speed anymore try move vdd
    ------- TIP: pciE speeds above 120mhz cause the onboard sata controller and other onboard periphals to fail
    ------- TIP: appearently the pci/pciE lock only starts to work at 167fsb+.
    ------------. if you have a problem at lower fsb speeds try 167fsb
    ------- TIP: the max fsb you can boot with is usually lower than the max fsb you can run in windows!
    ------------. thats why clockgen is very important on this board to increase the fsb in windows
    ------------. pentium m cpus initially boot with the default multiplier and fsb, then apply the fsb set in bios and then the multiplier set in bios.
    ------------. this means we cant boot at high fsbs because the stock multiplier with the increased fsb results in a huge cpu clock
    ------------. wich makes the cou crash the cpu, even though this high clockspeed is only applied for a fraction of a second.
    ------- TIP: set the jumpers on your ct479 to 100fsb wich should give you a 10-20mhz fsb boost
    ------- TIP: booting with 200fsb and increasing the fsb in windows with clockgen seems to give the best performence.
    ------------. once we boot with over 200fsb the bios probably automatically choses more relaxed advanced memory and chipset timings.
    ---- TIP: always backup your hdd if your ocing so you dont lose anything if the hdd corrupts
    ------- TIP: when you reboot the board it sometimes shuts itself down completely and then powers itself on again after a few seconds.
    ------------. dont worry, this is normal




    Links:

    main P4gd1 site
    asus' main site about the board with all specs and memory compability list

    P4gd1 downloads
    asus' download site for p4gd1 drivers, bios files and tools.



    Downloads:

    P4GD1 XS Edition Software Pack!
    since the asus servers hover between 10 and 20kb/s for download speeds i put all important stuff into one folder and zipped and hosted it at filefactory.com. click HERE if you have a problem downloading from filefactory.
    its 97mb in size (dont worry, its a 200kb/s download) and contains the following bios files apps and drivers:

    /BIOS/1006-final
    /BIOS/1007-beta2

    /DRIVER/Intel Chipset Drivers (default)
    /DRIVER/Intel software Installation Utility V6.0.1.1002
    /DRIVER/ITE IDE Controller
    /DRIVER/Marvell Gigabit Lan 8053
    /DRIVER/Marvell Gigabit Lan 8053 Open Source
    /DRIVER/Marvell Gigabit Lan 8053-VCT
    /DRIVER/Realtec alc861-5027
    /DRIVER/USB 2.0 Driver

    /TOOLS/915 Tweaker-beta
    /TOOLS/AMI BIOS Editor
    /TOOLS/Asus AIBooster 20035
    /TOOLS/Asus Bios Image Editor
    /TOOLS/Asus Probe
    /TOOLS/Asus Update
    /TOOLS/ClockGen for 954119
    /TOOLS/ClockGen for 954123
    /TOOLS/CPUZ-131
    /TOOLS/CrystalCPUID46
    /TOOLS/EIST Tool

    clockgen for the 954119 pll
    originally for the 954119 pll on the P5gd1 and other boards
    it also works for the 954123 pll though.
    *reported pci and pciE frequencies are wrong
    *clocks get applied faster than with the 954123 clockgen
    *Dont use CPUZ and Clockgen at the same time, Both will show increases System speeds though the system remains at the same clockspeed.
    If you overclock your system beyond its stable limit with both apps running and then reboot your memory may die! (happened to me 2 times)

    clockgen for the 954123 pll on the P4gd1
    *frequencies are applied slower than with the 954119 clockgen
    *correctly reports pci and pciE speeds and lets you change them
    to change the pci/pciE frequency click on the clock showing the current pci/pciE speed.
    *Dont use CPUZ and Clockgen at the same time, Both will show increases System speeds though the system remains at the same clockspeed.
    If you overclock your system beyond its stable limit with both apps running and then reboot your memory may die! (happened to me 2 times)

    915 tweaker beta
    allows you to change the basic memory timings in windows
    *cas doesnt seem to work
    *15.6µs was the fastest RMS setting for me (bh5 at 100-150mhz)
    *dont set RMS to "disabled", it will cause heavy memory corruption and might kill you windows install.

    EIST tool
    allows you to change multipliers and vcore in windows
    *the vcore option doesnt work for me
    *higher multpilers than the default one are locked like on a64 chips
    (unless you got an ES chip )

    latest version of CrystalCPUID
    same as the EIST tool but has some extra settings and it displays the cpu and system details like cpuz
    *to change multipliers go to file->multiplier management, set the multi you like, hit F3 to enable multiplier adjustments, hit F4 to see the multiplier and cpu frequency change (main window doesnt update the cpu speed automatically when chaning settings)
    *vcore settings dont seem to work

    AMI bios editor
    seems to work with P4gd1 bios files, but i have no idea how to use it
    *put your bios file of choice to the root directory of your drive, for example c:
    to load the bios enter c:/xxx.rom

    1006 bios
    the best bios for the P4gd1 i tried so far

    1007 beta2 bios
    this bios doesnt work with pentium M cpus!
    so dont flash it unless you have a prescott or northwood cpu

    SuperPI mod 1.4
    Superpi is a great tool to find the maximum benchable speed quickly.
    superpi 1m is a fast indicator or system stability, if it passes it doesnt mean your system is stable, but indicates the speed is most likely stable enough to run a few other benchmarks like 3dmark at a few (20-50) mhz less.
    superpi sometimes corrupts and keeps giving you an error message, if this happens just extract the files from the zip into the superpi folder again.

    MemTest 86 v3.2 Floppy Installer
    This tool is THE ultimate memory stress and stability test available.
    Test 5 and 8 are the most preferred tests as they quickly find memory instabilities. MemTest 86 cant run under Windows or DOS, you have to boot straight to MemTest so it has full access to the memory. There are 3 ways, boot from BIOS image, wich needs a modded BIOS and unfortunately isnt available for the P4gd1 (yet), boot from a floppy or boot from a cd.

    This is the Floppy Installer wich will extract all files necessary to boot to MemTest 86 from a floppy disk. format a floppy disk and use this installer to create a bootable MemTest 86 floppy disk. then reboot, enter the floppy disk and select to boot from floppy in bios.

    MemTest 86 v3.2 Bootable CD iso
    This tool is THE ultimate memory stress and stability test available.
    Test 5 and 8 are the most preferred tests as they quickly find memory instabilities. MemTest 86 cant run under Windows or DOS, you have to boot straight to MemTest so it has full access to the memory. There are 3 ways, boot from BIOS image, wich needs a modded BIOS and unfortunately isnt available for the P4gd1 (yet), boot from a floppy or boot from a cd.

    This is the CD iso you need to create a CD to boot to MemTest 86.
    just write the iso onto a CD, make sure you select "create a bootable CD". then reboot and select to boot from your CD drive in bios, put the CD into your CD drive and reboot.




    .
    ----
    -------------------------------------------------------------------------
    hey vince, did you ever try the increased fsb ermination voltage?
    does it help any?
    are there vcore and vdd mods already?

    i thougt about making a new thread about tweaking the board as the p4gd1 community is jumping for 10 to 50 now
    can you help me a little with that thread by posting your expiriences how to tweak the board?

  2. #2
    I would be glad to contribute whatever I can..
    If you want to cut and paste this or I can retype in new thread more organized.

    I was able to run the board stable with bh5 as tight as it can go up to 275+ without having to adjust fsb term voltage. Haven't needed to mess with that setting yet.
    Chipset mod is an absolute must. If you check with a dmm, you will find that chip voltage is set alot lower than setting in bios. On my p4gd1 I have here, if I use the 1.6v setting in bios..and check with a dmm at appropriate mosfet location, the chip volts would read 1.4x volts. Vcore mod is easy and a must. There is a vdd mod, but I just use my trusty booster as it works perfectly up to 3.6v actual. I see alot of guys having difficulties getting high fsb with this mobo. The trick is to boot at a certain fsb and have pci-e set properly. Then when you get into windows, you can hit your target fsb using clockgen.
    For example, with 780m and target fsb of 272 in windows for benching this is what i do:
    I set fsb in bios to 170-175(actual number depends on what you target fsb in windows will be) and set pci-e to 120-125(again, number depends on what your "target" fsb in windows will be).
    Ok, so I boot at 175fsb, 125 pci-e, and give chipset 1.85v. When I get into
    windows, I can now use cpuid and clockgen to adjust my multiplier to 14x and then crank the fsb using clockgen.
    If you do not do it like I described, you get hardlocks in windows as soon as you set highe fsb with clockgen. Ignore the pci-e freq reading in clockgen as it is wrong.

    I think this should halp about 99.9% of you that are having problems getting high fsb. I will think of some other stuff in abit...
    Last edited by k|ngp|n; 12-02-2005 at 11:58 AM.

  3. #3
    Xtreme Legend
    Join Date
    Jul 2004
    Location
    France
    Posts
    354
    Quote Originally Posted by saaya
    and about killing the memory, well my banias couldnt get over 150fsb for some reason, 150fsb was stable, 150.5fsb would freeze the system...

    when i had cpuz and clockgen running i could get over 150fsb... or so i thought
    actually the seed was still the same. when i closed cpuz and restarted it it was showing 150fs again... i rebooted and thats where my memory died.
    no idea how this happened, franck has no idea either...

    i tried the same thing again and again my memory died, so it wasnt a coincidence...
    I'm sorry for your memory, but IMO the cause is your mainboard.
    I don't think a memory stick can die just because of a too high frequency.
    A chip physical destruction always results from a voltage peek, AFAIK.

    Of course, as I explained in the PM, the o/c indirectly implies a peek in the mosfet stage, but only a faulty board can explain the memory destruction. What do you think ?
    Last edited by cpuz; 12-01-2005 at 06:41 AM.

  4. #4
    xtreme energy
    Join Date
    Oct 2004
    Location
    Europe, Latvia
    Posts
    4,145
    OK we know how to calculate PCI-E freq (pci-e * 2.2 = fsb). But how do you calculate FSB that you set in bios? Or any value from 167-200 is fine? Somehow I can't get stable system >250. Although I know my RAM is stable at 260+ speeds. Low chipset voltage is what I suspect as I haven't modded it yet. Also is there really a need for Vcore mod? Isn't 1.7V enough considering it overvolts by ~0.08V?
    Last edited by kiwi; 12-01-2005 at 06:45 AM.
    ...

  5. #5
    Xtreme Legend
    Join Date
    Jul 2004
    Location
    France
    Posts
    354
    This version of ClockGen will report the correct PCIE frequency :
    http://www.cpuid.com/download/ClockGen.zip

    Just choose "ICS 954123" in the PLL list ("PLL Setup" button).

  6. #6
    Quote Originally Posted by kiwi
    OK we know how to calculate PCI-E freq (pci-e * 2.2 = fsb). But how do you calculate FSB that you set in bios? Or any value from 167-200 is fine? Somehow I can't get stable system >250. Although I know my RAM is stable at 260+ speeds. Low chipset voltage is what I suspect as I haven't modded it yet.
    That formula is not really what I go by kiwi. YOu need to narrow down what is causing the lockups first. Start with your fsb at 170, and pci-e to 120.
    Set pci frequency setting to 33.33. Boot into windows with those settings, and then try to hit 250+ on your ram with clockgen. OF course, make sure your vdimm is where it needs to be for the ram to run at 250+.
    If you get an instant lock when you set 250 with clockgen...it is more than likely your chipset volts to low. CHeck with a dmm at the appropriate mosfet location to see what your getting. I bet it's way low.
    Without 1.75v+ on the chipset, I didnt get too far and had issues with locks setting high fsb. I would do the chipset mod and set to 1.75v and see where that gets you. I will take some close up pics of my board and mods and I will post them up in asec...If you still get the locks after you do the chip mod, go back into bios and boot at 171fsb, 121 pci-e. If that doesnt work, try 172,122. Then 173,123...and so on. Keep in mind, all of my experience with the board is with a 780m...this is what works for that chip no doubt.
    Last edited by k|ngp|n; 12-02-2005 at 05:05 AM.

  7. #7
    Quote Originally Posted by cpuz
    This version of ClockGen will report the correct PCIE frequency :
    http://www.cpuid.com/download/ClockGen.zip

    Just choose "ICS 954123" in the PLL list ("PLL Setup" button).
    Thanx for the info and your great work as always!!
    I am very appreciative of your oc'ingutlilites

  8. #8
    Chip mod 50kvr to 2nd leg from top on the left side of chip. Red circle is where to measure:


    My vcore mod:


    close up of vreg to mod:
    pin 8 to any ground.


    3.3v kyosen rail mod works great on this board for higher vdimm.
    Here is a pic courtesy of kamerat:
    Last edited by k|ngp|n; 12-01-2005 at 07:31 AM.

  9. #9
    silver wall jumper X
    Join Date
    Apr 2005
    Location
    Las Vegas
    Posts
    1,579
    Quote Originally Posted by cirthix
    hey everybody... i'm looking at 11 P4GD1s droolllll. the boards came in today. unfortunately, the shipping bios does not support the adapter and i dont have a 533/800fsb p4 to flash with. good news is that my local pc shop owner owes me a favor and said i could bring a board in (he doesnt know there are 11, so i might hotflash the rest and not burden him.. or i could just push my luck) and flash it using one of his cpus.

    more good news is the atx power plug, its 24 pin, but has the little nub all the way along it, so 20 pin connectors will have no problem clipping on.

    more good news: 2 people dropped today, the first 2 to pm me gets em

    bad news: due to shipping costing about 10 bucks per board more than what we expected, the costs of a board is now 150usd shipped. sorry for the increase, but shipping was just more than we expected (and paypal fees too, grr).
    AWESOME!!! they arrived! you gots PM

    Where shall we send payment to?
    Last edited by mike; 12-01-2005 at 07:34 AM.

  10. #10
    xtreme energy
    Join Date
    Oct 2004
    Location
    Europe, Latvia
    Posts
    4,145
    Thanks k|ngp|n, I will check that

    Quote Originally Posted by k|ngp|n
    Chip mod 50kvr to 2nd leg from top on the left side of chip.
    50? Or 10K like in this picture:
    ...

  11. #11
    I use 50k.

  12. #12
    Xtreme X.I.P.
    Join Date
    Nov 2002
    Location
    Shipai
    Posts
    31,147
    nice!

    vince, or anybody else, did you guys manage to come up with a way to set a different divider than 1:1 or 5:4?

  13. #13
    Xtreme X.I.P.
    Join Date
    Nov 2002
    Location
    Shipai
    Posts
    31,147
    Quote Originally Posted by cpuz
    I'm sorry for your memory, but IMO the cause is your mainboard.
    I don't think a memory stick can die just because of a too high frequency.
    A chip physical destruction always results from a voltage peek, AFAIK.

    Of course, as I explained in the PM, the o/c indirectly implies a peek in the mosfet stage, but only a faulty board can explain the memory destruction. What do you think ?
    no idea... but i dont think it was a voltage spike... vdimm was only 2.9v and seemed to be pretty stable... i only used 1 256mb stick, so there wass no load on the vdimm mosfets really, thinking this board supports at least 2 1gb sticks witch puts the vdim mosfets on 8x the load...

    i heard that memory can die if you keep running it at too high speeds where its not stable, and i heard that videocard memory can get damaged or die when you oc it too far and its showing artifacts.
    this never happened to me, but those are other scenarious i heard of that involve dead memory without a voltage spike.

    maybe i should contact asus and ask them about this...

  14. #14
    xtreme energy
    Join Date
    Oct 2004
    Location
    Europe, Latvia
    Posts
    4,145
    I don't think that is possible unless someone mods bios or writes some windows tool (like A64 tweaker for amd) to change divider.

    I would rather see bios where I can change multiplier, don't you guys agree
    ...

  15. #15
    Xtreme X.I.P.
    Join Date
    Nov 2002
    Location
    Shipai
    Posts
    31,147
    yeah... and couldnt it be that the options are already there but hidden?
    where are those bios modders when you need them

    tictac still owes me a favour, but its hard to get ahold of him theese days...

  16. #16
    xtreme energy
    Join Date
    Oct 2004
    Location
    Europe, Latvia
    Posts
    4,145
    I remember I was using this tool to unhide some AMI bios features (that was some ASrock board). However, it crashes every time I am trying to open bios file. Maybe someone else should try.

    More info here:
    http://www.dstyles.de/bios/en/
    ...

  17. #17
    Xtreme X.I.P.
    Join Date
    Nov 2002
    Location
    Shipai
    Posts
    31,147
    yepp, same here, it crashes... this one works, but i have no idea how to use it

    hmmmm where can i host that damn file? :P
    here it is:
    http://www.filefactory.com/get/f.php...22a7d2e8820a1d

    bios 1006:
    http://www.filefactory.com/get/f.php...af020e1d3cfab9

  18. #18
    Xtreme Member
    Join Date
    Oct 2002
    Location
    Oslo, Norway
    Posts
    294
    Quote Originally Posted by kiwi
    50? Or 10K like in this picture:
    http://home.no.net/comerade/pc/p4gd1_chipset.jpg
    10k gives higher voltage when pot is set to maximum resistance, but it's easier to fine tune the voltage with a 10k one.
    Satan: Opteron 165 - TT Big Typhoon - DFI Lanparty NF4 SLI-DR - OCZ EB PC3500 2x512MB - Asus Radeon X800 XT - Raptor 74GB x2 RAID0
    Slepbar (Lenovo 0761): Pentium M 735 (Celeron M 380 @ 2,13GHz) - DDR2 2x512MB - GMA 900
    3DMark2001: 40392 - Super PI: 23s

  19. #19
    Xtreme Addict
    Join Date
    Nov 2004
    Location
    DK
    Posts
    1,026
    Already a lot of info condensed in the "Dothan - Howto and Fixes" thread.

    Feel free to use it...

  20. #20
    Xtreme Member
    Join Date
    Dec 2004
    Location
    france
    Posts
    371
    there is a problem if i use mod +3.3 with psu at 3.85 on this board ?
    Signature removed.
    Links in signatures to unapproved and/or unaffiliated sites is not allowed.
    Re-read the Forums Policies & Guidelines.

    Signature edited by IFMU

  21. #21
    Xtreme X.I.P.
    Join Date
    Nov 2002
    Location
    Shipai
    Posts
    31,147
    Quote Originally Posted by Kamerat
    10k gives higher voltage when pot is set to maximum resistance, but it's easier to fine tune the voltage with a 10k one.
    you mean 50k gives a higher voltage but 10k is easier to finetune, right?

    Quote Originally Posted by stryg
    Already a lot of info condensed in the "Dothan - Howto and Fixes" thread.

    Feel free to use it...
    thx
    will go and collect infos this weekend

    Quote Originally Posted by skirms_fr
    there is a problem if i use mod +3.3 with psu at 3.85 on this board ?
    what happened?
    did it work with the rail tunes to less volts?

  22. #22
    Xtreme Member
    Join Date
    Aug 2005
    Posts
    244

    Thumbs up

    Thanks
    Last edited by 16floz470ml; 12-02-2005 at 12:29 PM.
    i7 920...Intel DX58SO
    12gb Patriot Viper Xtreme
    Palit GTX 560 ti 2gb SLI
    Antec Quattro 1000...Twelve Hundred Case
    Intel X25-M 80gb x 4 Raid 0...Hitachi 7K3000 2tb
    Sony GDM-FW900...Z-5500...X-Fi Forte

  23. #23
    Xtreme Member
    Join Date
    Oct 2002
    Location
    Oslo, Norway
    Posts
    294
    Quote Originally Posted by saaya
    you mean 50k gives a higher voltage but 10k is easier to finetune, right?
    No, both 50k and 10k can give deadly voltages. But the 10k have a higher starting voltage when set to 10k than a 50k set 50k.
    Satan: Opteron 165 - TT Big Typhoon - DFI Lanparty NF4 SLI-DR - OCZ EB PC3500 2x512MB - Asus Radeon X800 XT - Raptor 74GB x2 RAID0
    Slepbar (Lenovo 0761): Pentium M 735 (Celeron M 380 @ 2,13GHz) - DDR2 2x512MB - GMA 900
    3DMark2001: 40392 - Super PI: 23s

  24. #24
    xtreme energy
    Join Date
    Oct 2004
    Location
    Europe, Latvia
    Posts
    4,145
    Quote Originally Posted by Kamerat
    No, both 50k and 10k can give deadly voltages. But the 10k have a higher starting voltage when set to 10k than a 50k set 50k.
    yes, you usually lower resistance for more volts. So basically 20K, 30K... also can be used. It would even be safer depending on how much volts does 10K give you.
    ...

  25. #25
    Xtreme Member
    Join Date
    Sep 2004
    Location
    Brazil
    Posts
    401
    Quote Originally Posted by kiwi
    yes, you usually lower resistance for more volts. So basically 20K, 30K... also can be used. It would even be safer depending on how much volts does 10K give you.

    Great thread Saaya , good to know how you guys are doing with this board
    E6600@3.3Ghz
    P5B-Dlx ( 0711 bios )
    BBA X1900XTX
    4x512 Corsair 5400UL


    ----------------------------------------------------
    "They couldn't hit an elephant at
    this distance" (last words of Gen.
    John Sedgwick, Spotsylvania 1864) "

    ----------------------------------------------------

Page 1 of 13 123411 ... LastLast

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •