Okay I'll take a different direction. When I connect my external DVD drive via USB, the bootup process is very much longer. I think it's because the external drive isn't responding quite the way it's supposed to, and the laptop blocks till the drive responds.
If there are issues with the built-in drive, it might be possible that the drive fails to respond sufficiently quickly. To test, we can either change the boot order, or disconnect the drive. Either way, should be rather easy to test this...
-
TofuTurkey Married a Champagne Mango
-
I run my machine without a DVD/BluRay drive. The drive failed a while back and I haven't tried to get it fixed yet so I disconnected it.
No difference either way.
I'm very comfortable my issue was constrained to the specific drive and not caused by the laptop itself. I exchanged the drive with the one in my older Z. The old drive worked fine and the new drive didn't work in the old Z either. -
Hey Zoinks, do you have the issues when coming from shutdown, reboot, hibernation? do the timeouts differ?
I noticed that there is a reduced timeout when coming from hibernation. -
Seems the same to me.
-
In case it helps, I did a clean install JBOD a few months back and have no blinking problems.
-
I'm going to get back into troubleshooting this over the weekend.
This will include a call into Sony support. -
TofuTurkey Married a Champagne Mango
There's something else to try: I did the conversion of JBOD to RAID0 via the Intel Storage Manager, and the boot up was ok. I'm wondering if clean install directly into a RAID-ed disk results in that problem.
-
Tried that. The slow boot behavior remained.
-
ZoinksS2k
I can suggest you to create topic with this problem on any sony repair service forum.
Maybe repair engineers can help you. -
I'm working on the problem right now. I've posted in a few other places to get help but don't know of a Sony-owned forum.
Do you have a link? -
*sigh*
-
Bill
have almost exactly the same issue on my Z11 -- my system is the i7 512G 6G, with Verizon card
I was only able to get it with the trialware, initial boot ups after bios screen was about 3 blinks and almost immediate starting windows message.
did restore selecting the option not to install extra software, which worked well
then I got about 17 blinks and then something like 12 seconds of black until windows starts.
to make it more curious, I have another Z11 system bought about 6 weeks before. it was CTO and was i7, 128G 4G. did a restore of that system and it did not change the boot up timing, 3 blinks and pretty much right to starting windows.
so I have one system where the restore did not hurt, ( 128G ssd) and the other one where it wound up like yours (512G SSD)
I was on the phone with sony, and said that the restore disks do not return the system to factory shipped condition. the level 2 guy admitted there is an issue. He sent me official restore disks (BTW there were 4 for some reason) did the restore with them and no difference was noted.
I will be back to them to follow up. I will not be happy just to send the system to them, because it will wind up in the same state with the problem not resolved. I agree with you a restore disk should restore! -
Thanks for the info. Looks like Sony is at least aware of the issue.
-
They must have a build manager system to create this image and at some point I am sure that there is a file , a registry entry or something modified that does not get done when a restore dvd is made or if you use the recovery partition.
the fact that it is reported that a plain win 7 install does not solve the problem makes me think think that there is something included on a factory machine, that windows or the published drivers from Sony do not include.
once you do a restore, the setting or file is lost.
The fact that my 128G system did not suffer from being restored, and my 512G system did, means that someone in the software group that does these things must have gone out of their way to make a fix or patch to the system, but somehow it did not make it onto the software distribution or recovery partition. So this is an issue with configuration management of the software and is not really a technical issue. -
Agreed.
The issue does not occur with Ubuntu, Windows XP or FreeBSD. Seems to only affect Windows Vista and Windows 7, and only when a RAID 0 volume is used. I tried both in x32 and x64 flavors. -
This post may add to the frustration and bewilderment of those experiencing unusually long boot times with their VPCZ11s.
After performing a VAIO Care Recover of drive C: on a two-day old VPCZ118GX/S (3 x 128 GB SSDs) I noticed that the cursor blinked 17 or 18 times from the BIOS screen until the Windows start screen appeared. After grappling for some time with the delayed boot issue I exchanged the unit for a new one convinced that the issue was caused by restoring my notebook to "original factory condition" and that a unit with a factory-fresh image would not exhibit the behavior (as long as I did not perform a restore or a clean install).
I was mistaken. During the first two days I got 10 or 11 blinks. Then it crept to 14. Now I am up to 17 or 18 blinks the same number of blinks as with the unit that I returned.
Why the creeping boot delay? What is causing this? Is it going to get progressively worse?
For those who may be able to make sense of it, I offer the following cut and paste of my log.
First boot from Windows 7 (64-bit) Repair Disk
Command Prompt
DISKPART
SELECT DISK 0
LIST PARTITION
Partition ### Type Size Offset
Partition 1 Recovery 8 GB 1024 KB
Partition 2 Primary 100 MB 8 GB
Partition 3 Primary 348 GB 8 GB
BCDEDIT
Windows Boot Manager
identifier {bootmgr}
device partition=C:
description Windows Boot Manager
locale en-US
inherit {globalsettings}
default {default}
resume object {a9d8649e-3667-11df-8931-001dbac1843c}
display order {default}
toolsdisplayorder {memdiag}
timeout 30
Windows Boot Loader
Identifier <default}
device partition=D:
path \Windows\system32\winload.exe
description Windows 7
locale en-US
inherit {bootloadersettings}
osdevice partition=D:
systemroot \Windows
resumeobject {a9d8649e-3667-11df-8931-001dbac1843c}
nx OptIn
Second boot from the factory-installed SSD array
Command Prompt
DISKPART
SELECT DISK 0
LIST PARTITION
Partition ### Type Size Offset
Partition 1 Recovery 8 GB 1024 KB
Partition 2 Primary 100 MB 8 GB
Partition 3 Primary 348 GB 8 GB
BCDEDIT
Windows Boot Manager
identifier {bootmgr}
device partition=\DeviceHarddiskVolume2
description Windows Boot Manager
locale en-US
inherit {globalsettings}
default {current}
resume object {a9d8649e-3667-11df-8931-001dbac1843c}
display order {current}
toolsdisplayorder {memdiag}
timeout 30
Windows Boot Loader
Identifier {current}
device partition=C:
path \Windows\system32\winload.exe
description Windows 7
locale en-US
inherit {bootloadersettings}
recoverysequence {a9d8649e-3667-11df-8931-001dbac1843c}
recoveryenabled Yes
osdevice partition=C:
systemroot \Windows
resumeobject {a9d8649e-3667-11df-8931-001dbac1843c}
nx OptIn
While conducting research I came across the marketing specifications sheet for the VPCZ118GX/S which states that the averge boot time is 60 seconds " ased on original factory image and will vary with product configuration changes and additional software added to the hard drive."
Using the stopwatch in my BlackBerry I timed the total boot time of my VPCZ118GX/S in 54.88 seconds (from pressing the power button until getting a normal Windows 7 cursor). Therefore, Sony is likely to deem this as "normal behavior" despite the fact that other end-users with SSD-equipped notebooks are reporting 20-second boot times.
I will now go over the posts on this thread because it appears that the length of the blinking cursor delay may depend, at least in part, on the number of SSDs in the RAID configuration. -
TofuTurkey Married a Champagne Mango
Edit: Oops one minute late... -
-
Well if it is a creeping delay, it does not explain why my older 128G ssd i7 machine is booting exactly the same as when we got it, (its been a couple of months now) and did not suffer after a restore.
Even if there is some kind of creeping going on, something changed suddenly in these units after the restore. It could be two issues that have some symptoms in common -
Nevertheless, I think this an issue caused by Microsoft, not Sony - so our conversational partner might be Microsoft. -
-
-
-
-
-
-
-
-
I'm working on this very thing. I hope to have something done today.
I've installed the Windows performance debugger as well. -
One thing is sure, I dont think a solution will be forthcoming from Sony any time soon -
I'm currently dorking around with the windows boot loader.
I don't remember if I asked this before or if anyone answered, but I need somebody with a fast-booting machine to dump their BCD config.
Open an elevated command prompt and type (you can copy/paste this):
Code:c:\windows\system32\bcdedit.exe /enum /v
-
TofuTurkey Married a Champagne Mango
1. A place to put the config/menu files. I don't know if GRUB can pick these up from an NTSF partition, otherwise an EXT partition will have to be created.
2. Auto selection and loading of the Windows OS. If the delay is set to 0, does it mean GRUB waits indefinitely for user input?
?? -
We should be able to dump the grub loader into the System Reserved partition, replacing the BOOTMGR since it can work on NTFS if the MBR is changed.
I haven't quite given up on getting the windows loader to work, but I'm getting darn close.
If you guys are interested in helping, I think WinGRUB will do the trick. I'll be trying later. -
But that said, I don't think it will help at all, because grub has to chain the Windows bootloader, and it's when the Windows bootloader tries to enumerate the RAID that the slowness occurs. You just add one more step in front of the delay. -
TofuTurkey Married a Champagne Mango
-
FWIW, I used bcdedit /enum all to make a text file of my boot configuration.
it may ( or may not) be interesting to compare this to a machine that is booting fast.. guess on the 0.01% chance it is something obvious.......
Windows Boot Manager
--------------------
identifier {bootmgr}
device partition=C:
description Windows Boot Manager
locale en-US
inherit {globalsettings}
default {default}
resumeobject {36096564-6bb0-11df-bb24-85836f647e1a}
displayorder {default}
toolsdisplayorder {memdiag}
timeout 30
Windows Boot Loader
-------------------
identifier {default}
device partition=D:
path \Windows\system32\winload.exe
description Windows 7
locale en-US
inherit {bootloadersettings}
recoverysequence {36096566-6bb0-11df-bb24-85836f647e1a}
recoveryenabled Yes
osdevice partition=D:
systemroot \Windows
resumeobject {36096564-6bb0-11df-bb24-85836f647e1a}
nx OptIn
Windows Boot Loader
-------------------
identifier {36096566-6bb0-11df-bb24-85836f647e1a}
device ramdisk=[E:]\Recovery\WindowsRE\Winre.wim,{36096567-6bb0-11df-bb24-85836f647e1a}
path \windows\system32\winload.exe
description Windows Recovery Environment
inherit {bootloadersettings}
osdevice ramdisk=[E:]\Recovery\WindowsRE\Winre.wim,{36096567-6bb0-11df-bb24-85836f647e1a}
systemroot \windows
nx OptIn
winpe Yes
Resume from Hibernate
---------------------
identifier {36096564-6bb0-11df-bb24-85836f647e1a}
device partition=D:
path \Windows\system32\winresume.exe
description Windows Resume Application
locale en-US
inherit {resumeloadersettings}
filedevice partition=D:
filepath \hiberfil.sys
debugoptionenabled No
Windows Memory Tester
---------------------
identifier {memdiag}
device partition=C:
path \boot\memtest.exe
description Windows Memory Diagnostic
locale en-US
inherit {globalsettings}
badmemoryaccess Yes
EMS Settings
------------
identifier {emssettings}
bootems Yes
Debugger Settings
-----------------
identifier {dbgsettings}
debugtype Serial
debugport 1
baudrate 115200
RAM Defects
-----------
identifier {badmemory}
Global Settings
---------------
identifier {globalsettings}
inherit {dbgsettings}
{emssettings}
{badmemory}
Boot Loader Settings
--------------------
identifier {bootloadersettings}
inherit {globalsettings}
{hypervisorsettings}
Hypervisor Settings
-------------------
identifier {hypervisorsettings}
hypervisordebugtype Serial
hypervisordebugport 1
hypervisorbaudrate 115200
Resume Loader Settings
----------------------
identifier {resumeloadersettings}
inherit {globalsettings}
Device options
--------------
identifier {36096567-6bb0-11df-bb24-85836f647e1a}
description Ramdisk Options
ramdisksdidevice partition=E:
ramdisksdipath \Recovery\WindowsRE\boot.sdi -
esh, I actually wanted the guids which is why I added the /v switch.
Anyhoo, I don't think we will get anything from changing the BCD info. I tried all the obvious stuff. -
I've got my machine dual booting with Ubuntu (needed to do this anyway).
Who was saying it made their system boot faster? I am not seeing it.
Arth1 was right, it just defers the delay. It did not make it any slower, but with a 2 second choice for OS, the boot times were practically the same as Win7 by itself. I guess there was a small benefit somewhere.
I'm about to wash my hands of this, at least for a while. Spending 2 hours to save 10-15 seconds on boot-up isn't worth it anymore.
Here are some rough times I collected:
Drive Config - JBOD
OS - Windows 7 x64 (fresh install)
Power On to Desktop - 33.8 seconds
Drive Config - JBOD on OS, RAID0 on remaining three drives
OS - Windows 7 x64 (fresh install)
Power On to Desktop - 38 seconds
Drive Config - RAID0
OS - Windows 7 x64 (fresh install)
Power On to Desktop - 49 seconds
Drive Config - RAID0
OS - Windows 7 x64 (fresh install) and Ubuntu 10.04 (grub)
Power On to Desktop - 50 seconds
I'm going to the pool -
I am immediately at the GRUB console and the windows startup logo appears immediately after chainloading the boot loader. -
thanks -
The problem isn't with grub, but when winload takes over.
I think you aren't seeing the problem because your machine doesn't have the problem (thankfully).
One article I found shows why this behavior happens, but only in Windows 2000 and with NTLDR. I was not able to find the equivalent setting with the newer BOOTMGR that Vista and Win 7 use.
Windows May Use Signature() Syntax in the Boot.ini File
The issue may be back in Sony's court with a possible defect in the BIOS with how it presents INT13 and INT14 calls when setup in RAID mode. Then again, I could be wrong. If this were the case, it would happen all the time. -
I couldn't leave well enough alone...
I took off the keyboard and physically disconnected one of the SSD drives, or sandwiches as I like to call them, and configured RAID with just a single drive. As you know, each "drive" is actually two SSD's. One on each side.
Configs/results:
- Single drive sandwich, RAID0, fast. (33 sec)
- Installed the RST 9.6 and rebooted again, fast. (33 sec)
- Reconnected the other drive sandwich and left them JBOD, fast. (34 sec)
- Reconfigured all drives for RAID in the Intel Option ROM, installed Windows. Slow. (50 sec)
- Repeated steps 1 through to step 3, fast (32 sec)
- Expanded RAID volume from the RST app.... fast. (34 sec)
So there you have it. Boot times are reasonable once again.
What's weird about this is that I reconfigured the RAID volume to use only one drive (ports 0 and 2) from the Intel Option ROM but without disconnecting the second sandwich. Boot times were always slow.
Not sure why, but building the RAID volume this way helped. The good news is that HD Tune and CrystalMark both show that performance is normal, meaning nothing was lost in rebuilding the stripe in this manner.
Word of warning before you try this, be very careful. The connectors on the SSD are fragile. I'd recommend trying to config your system with drives 0/2 in RAID and 3/4 in JBOD first. I tried it and it didn't work but I could have messed something up. This is the safest option. -
Well so much for washing your hands, and going to the pool.
I don't have the system with me at this time (another location). I have a couple of questions
Does 9.6 give you more ability to configure the array than the stock RST?
the notes on 9.6 I find say it is for a desktop. Exactly which version are we talking about.
your last comment in the post-- are you saying this is an alternative to physically unplugging the drives?
Could it be that building the stripe this way where you expand it using RST, it behaves like a stripe of two logical disks on the part of the array you loaded, and by adding more disks it retains the original two stripe, and simply appends the extra drive(s) to what is already there, but now there is more space total.
the slow way it builds a stripe of 4 logical disks
I could believe that they do it that way in the factory, but did not get it right when the recovery disks are made. -
Like I said, I can't leave well enough alone
There are no differences in the Intel Control Panel between RST versions. I use 9.6, but the Sony-supplied version will give you the same options.
You can't add disks to a RAID volume from the BIOS screen during post (Intel Option ROM). You can do this from the Intel RST once you boot to Windows. You simply click "Add Disk" select the ones you want and go. It migrates the data as it expands the volume.
I'm not really sure what happens during the factory build. -
OK I was able to get it to work on my system too, but without cracking the Z open
1. deleted raid drives on intel raid bios
2. added the first 2 drives in the bios, leaving the other two alone
3. did normal restore, complete system, but opted out of all value added software, resulting in more or less a fresh start system
4. added as you suggest, the other two drives to the array using RST, RST rebuilds the array
boot time was 35 sec from the bios screen to the windows this is 20 seconds faster than before
Good work Z thanks for your great effort, I think this will help those who want to do a restore, but I still will follow up with sony so they just fix the restore capabilities so we don't have to do all this. Although they could just build it into the restore script
below are the before and after tests, they are pretty close, however once I extended the c drive with the unallocated space, the sequential read dropped by about 25%. (to maybe 450MB/sec) numbers were the close I am playing around with that
Before 4 disk -- standard build, 55 second boot time
After 4 disk, 2 raid0 build with 2 added with RST, 35 sec boot time. This did not include added volume space, after that the Seq read ws about 450 -
1. Download the latest version of grub4dos and grubinst
2. Copy the files grldr and menu.lst that come with grub4dos to the root of your system drive
3. Run grubinst from an elevated command promt
Code:grubinst.exe (hd0)
Code:timeout 0 title Windows 7 find --set-root --ignore-floppies /bootmgr chainloader /bootmgr
Code:boot\bootsect.exe /nt52 c:
Be sure to have a working installation dvd / stick ready in case you mess something up. The original windows boot loader is restored by
Code:boot\bootsect.exe /nt60 c: /mbr
There is another advantage in using GRUB even if you still experience a delay. GRUB can be forced into debug mode using the insert key (or the debug directive in menu.lst) - this might help to find out what is causing the delay. -
and your system SSD arrangement?
-
Code:DISKPART> list disk Datenträger ### Status Größe Frei Dyn GPT --------------- ------------- ------- ------- --- --- Datenträger 0 Online 238 GB 13 GB Datenträger 1 Kein Medium 0 B 0 B Datenträger 2 Kein Medium 0 B 0 B DISKPART> select disk 0 Datenträger 0 ist jetzt der gewählte Datenträger. DISKPART> list partition Partition ### Typ Größe Offset ------------- ---------------- ------- ------- Partition 1 Primär 224 GB 1024 KB DISKPART> list volume Volume ### Bst Bezeichnung DS Typ Größe Status Info ---------- --- ----------- ----- ---------- ------- --------- -------- Volume 0 G DVD-ROM 0 B Kein Medi Volume 1 D DVD-ROM 0 B Kein Medi Volume 2 C NTFS Partition 224 GB Fehlerfre System Volume 3 F Wechselmed 0 B Kein Medi Volume 4 E Wechselmed 0 B Kein Medi DISKPART>
-
Sony VPC-Z11 - Slow POST
Discussion in 'VAIO / Sony' started by ZoinksS2k, May 17, 2010.