I just have Vantage to go. Going to run this morning.
If it ain't broke, don't fix it :(
Plan to add one more Vertex 4 256GB in R0 just like to try with modded OROM before.
Replaced in Z77MX-D3H F14 bios ROM module 11.1.0.1413 to 11.5.0.1582. As well updated RST drivers in OS.
ASSSD went from 1100+ to 1030-1040.CDM read 500+ to 400+ as well 4K bit. Won't bother to add screens.
Was not so happy about performance drop. As not adding another dive yet though simply roll back changes, no big deal.
Left before OROM/RST, right after flashing back original bios and changing back RST drivers in OS from version v11.6.0.1030 to 11.0.0.1032.
http://www.upload.ee/image/2690862/O...V-rollback.jpg
So is there any trick I am missing about rollback? Any difference about specific sequence to install MB INF/RST? Cleaning registry? Something else beside clean install.
Like to get back missing 100MB/s in seq and 20MB/s in 4K writes until need TRIM in R0 :)
Exactly: You have to clean the registry from a faulty iaStorF entry, which has been forgotten to erase by the RST(e) drivers uninstall procedure.
You can easily verify, if you are effected by this issue, when you open the Device Manager and look into the driver details of your drives (right click > "Properties" > "Driver" > "Driver Details"). If you should find the iaStorF.sys there and within the Windows\system32\Drivers folder as well, you got the performance decrease by the SCSI filter driver iaStorF.sys, which is still active, but cannot really be used anymore.
Warning:
Don't try to delete the iaStorF.sys from the Drivers folder, because this will induce a BSOD at the next restart.
Solution:
Run the regedit command, open the HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Contro l\Class\{4D36E967-E325-11CE-BFC1-08002BE10318} or similar and delete the word "iaStorF".
After having restarted your PC everything should be fine again. You will even be able to delete the iaStorF.sys from the System32\Drivers folder.
No and no!
What I have written is only valid for users, who had switched back from any Intel RST(e) drivers v11.5 or v11.6 series to the "normal" RST drivers, which do their job with just 1 single driver named iaStor.sys.Quote:
Is this something we should be doing as standard procedure?
Oh crap i removed it. What does this do?
Do you mean the registry entry "iaStorF"?
If you mean the registry entry, it gives the command to use the SCSI filter driver named iaStorF.sys. This is a big mistake, because the iaStorF.sys is not part of the "normal" Intel RST driverpacks v10.x.x.xxxx, v11.x.x.xxxx or v12.x.x.xxxx and has not been designed by Intel for being used together with the RST driver named iaStor.sys.Quote:
What does it do!
how come iastorf was in my registry I've never used any rst e drivers
read latency is faster than the V4 and it has nearly twice the performance of the V4 in real world testing. The Vertex 4 isn't even in the same league. When in RAID 0 these drives really put the V4 in RAID 0 to shame. here are the access times in raid 0:
http://i876.photobucket.com/albums/a...ps3403ba22.png
http://i876.photobucket.com/albums/a...psfadbd62e.png
full reviews coming very soon
No, you should not remove the SCSI filter driver as long as you are running one of the RST drivers v11.5 or higher. Ma advice to remove the iaStorF entries within the registry was only valid for users, who previously has installed one of the RST(e) drivers v3.x.x.xxxx, v11.5.x.xxxx, 11.6.x.xxxx or 12.x.x.xxxx and have switched back to the 1-driver RST system.
My prognosis: Intel will stop or has already stopped the development of new drivers on the basis of the conservative 1 driver strategy. That means, that the v11.2 series drivers probably will be the last ones with an iaStor.sys as the only driver.
If I am right, we all will have to live with the additional SCSI filter driver, if we want to use new Intel AHCI or RAID drivers.
By the way: It was the SCSI filter driver iaStorF.sys, which brought Intel the break-through regarding Trim support within a RAID array.
but the 11.2 still allow pass through, and no scsi. i don't see the correlation. up until a month or so ago, i assumed the scsi driver was needed for sata raid pass through on the intel controller, but it appears to work without it, as under the 11.1/11.2?
i don't know which real world you speak of, but for half the storage (and a sandforce controlled drive to boot..arg), and at a $100 premium over a 256gb V4, plus questionable support (never heard of this 'brand'), it must be performing unmentionable acts for you, to praise it in such a manner...sorry, off topic...
here this will explain it better:
http://i876.photobucket.com/albums/a...ps15699449.png
Workstation is the same as a typical user environment. Notice TWICE the performance of a V4
http://i876.photobucket.com/albums/a...psbfc9b5b5.png
Drives filled 50% with data. Notice Vertex 4 scores 39,000 and the SuperSSpeed scores 69,000 one is a Ferrari the other a Buick
The company that you have never heard of is the largest distributor of Intel NAND flash in their region. They've been making flash based products for several years and products that we see as 'trusted' brands all of the time. They are just doing their own branded drives now. Also, the parent company is based in the US.
after seeing vantage tests give compressible results for supposed incompressible situations...i guess i don't value it as 'real world', though it's nice to see slc at such low prices. would be nice to see it in other ssds controlled by something other than sandforce, and as for the 'parent' company being in the US, that's great, but after going through many US based forums of other companies, i guess i would want to know a bit more re their support, especially with something 'new' like this coupling of slc on the sandforce targeted for consumer use in the US (i'll have to check them out), but we are way off topic at this point...
I've got way way off topic on this thread and apologize for that I made a new thread http://www.xtremesystems.org/forums/...60#post5138460 to discuss the SLC drives Verses any other drives.
Just today, I have new Intel® RAID ROM 11.6.0.1624 (not uploaded yet).
Supose to be 11.6.0.1702.
Anybody know about this?
Thanks to Stasio I'm running it now and it has provided better performance across the board. :yepp:
Intel® RAID ROM 11.6.0.1624
http://www.mediafire.com/?p48mqf64otwycyc
^
Woooo,nice. ;)
guys i need help
i just updated bios with new OROM and i made an image with acronis of my raid to another drive
now i erased my SSD's made the raid again and acronis doesnt see it and windows 7 install also ask me for drivers :( how can i recover my image now :(
thanks in advance
Sergio
I would try (if you have a spare hdd) to install a new OS to a spare disk and then install Acronis and try from there to restore the image...maybe this works...
can i flash the new rom on my p8z68-v pro/gen3???
how?
new OROM OUT ??? Intel® RAID ROM 11.6.0.1702
Modded Gigabyte x58 bioses:
http://forums.tweaktown.com/gigabyte...d-bios-61.html
Thanks Bill did anyone try it yet ?
TRIM w/ RAID0 seems a bit hit and miss for me
I've got an Asus Z77 board w/ 11.0 OROM, 2 x 256 Samsung 830's raid0, win7 x64 sp1
I installed 11.2 RST and confirmed trim was working using the hex method and left it at that..
Of course I couldn't let it be, when 11.6 came out I inserted the 11.6 OROM and reflashed the bios and installed 11.6 RST and went to test trim and found it didn't work
I reverted back to 11.0 OROM and 11.2 RST and tested TRIM again, seemed to work. That is until I tried a few hours later and it's stopped working altogether
I'm not sure if its just never been working and I've accidentally been pasting the sector in and leaving the 0 in there which I've spotted a couple of times, but this is starting to annoy me.
I tried 11.5 and still nothing
I'm going to try Windows 8 and if that's unsuccessful giving up on TRIM working for me, or perhaps with Samsung 830's it's not instant I don't know..
Whilst I don't disagree it's certainly not what I'm experiencing
On a fresh Windows 8 install w/ 11.6 it worked once, then when I attempt the exact same JPG which ends up on a different sector it doesn't clear the values
Same thing happens on Windows 7 now
On Windows 8 I've managed to repeat the hex test twice in a row where it clears the values. On the third and subsequent attempts the values don't clear
I'm not really sure whats going on with the underlying system and whether TRIM permanently or momentarily stops working. Or whether the hex test is an unreliable test method
Either way, I'm just going to assume that one successful test = TRIM working and leave it at that otherwise I'll end up forever trying to figure this out for something that still seems to be in its infancy
@ raab:
Switching back from the RST(e) drivers v11.5/11.6 to the "normal" RST driver iaStor.sys may give you some bad iaStorF residues in the registry. To verify it, open the "Drives" section of the Device Manager > "Driver details" and look, if the iaStorF.sys file is still listed. If yes, don't try to delete the iaStorF.sys from the Windows\System32\Drivers directory. Before you do that, you have to erase the wrong "iaStorF" entries within the registry.
I did clear the residue but still had the same results. Worked 1 or 2 times then not again. Might be a quirk with the Samsung 830's, who knows..
I've switched to Windows 8 now anyway where I'm having interesting results with UEFI driver vs Legacy OPROM being different versions? I've posted on the win-lite.de forums anyway
Intel SSD Toolbox 3.1.0 changelog lists "Trim on RAID 0 with RST 11.0 or greater"
But I think this means only the ability to manually trim.
Actually, you can use the Intel SSD Toolbox to set a TRIM schedule in Windows 7/8 (as well as doing a manual Trim).
Unfortunately if you are running your SSD's in RAID0 in Windows 8 (or Windows Server 2012) you cannot run TRIM at all.
Hopefully Intel will have support for Windows 8 soon.
Intel SSD Toolbox 3.1.1 has been released to fix a problem with detecting drive health and also the firmware on the 330.
this are my adatas on my M5F Z77
optimize is grayed...
http://i.imgur.com/tJNdi.png
With which IRST driver and Option ROM?
I tested with different IRST drivers and OROMs and found that Trim for RAID0 is only working on Windows 7 (despite what the Intel Download Centre seems to suggest).
Here are some screenies from my testing.
Windows 7 + Intel RST Driver 11.2.0.1006 + OROM 11.2.0.1527 = TRIM in RAID0 works
Windows 7 + Intel RST Driver 11.6.1.1002 + OROM 11.6.0.1702 = TRIM in RAID0 works
http://members.iinet.net.au/~michael...ptimize_Ok.png
Windows 8 + Intel RST Driver 11.2.0.1006 + OROM 11.2.0.1527 = TRIM in RAID0 not supported
Windows 8 + Intel RST Driver 11.6.1.1002 + OROM 11.6.0.1702 = TRIM in RAID0 not supported
http://members.iinet.net.au/~michael...o_Optimize.png
Windows 8 + Intel RST Driver 11.2.0.1006 + OROM 11.2.0.1527 + Toolbox v3.1.1 in Windows 7 Compatibility Mode = Error
Windows 8 + Intel RST Driver 11.6.1.1002 + OROM 11.6.0.1702 + Toolbox v3.1.1 in Windows 7 Compatibility Mode = Error
http://members.iinet.net.au/~michael..._Mode_Fail.png
You are absolutely wrong!
When the "Customize" button of the Intel SSD Toolbox is greyed out or the optimization doesn't have any effect, it only means, that the manually invoked TRIM feature of Intel's SSD Toolbox doesn't work with the related SSDs or Operating System, but it doesn't mean, that the permanently active "TRIM in RAID0" feature of the actual Intel RST drivers doesn't work.
Just to verify that again I just have redone the quick Hex Editor TRIM test on both Operating Systems Win7 and Win8 of my Intel Z77 chipset RAID0 system running the v11.6.2.1002 RST drivers and the v11.6.0.1702 RAID ROM.
Result: TRIM is active within the RAID0 on Windows 7 and Windows 8.
I see - so it's just the Intel SSD Toolbox that can't manually TRIM in Windows 8.
I decided to do the Hex editor test myself to confirm that TRIM is working, using Coup27's method;
I did not even have to wait for 10 mins, within 2 mins the file was gone (the offset that the gif file was located at contained all zeroes from that offset onwards) TRIM is definitely working in Windows 8, I am currently using driver 11.2.0.1006 + OROM 11.2.0.1527.Quote:
Originally Posted by Coup27
Exactly.
Here I have written a detailed tutorial about how to do it: http://www.station-drivers.com/forum...hp?f=32&t=4023Quote:
I decided to do the Hex editor test myself to confirm that TRIM is working, using Coup27's method;
If TRIM is active, the data are already erased within 1-2 seconds.Quote:
I did not even have to wait for 10 mins, within 2 mins the file was gone
Has anybody successfully gotten TRIM to work on a P8P67 R0 array, using a custom OROM 11.x BIOS?
I've read yays and nays - mostly nays. If not I see a P8Z77-V in my near future. Ugh.
better get that Z77 board because the yays are dreaming. BTW here is a review that shows RAID 0 TRIM working on both Windows 7 and Windows 8 and compares storage performance of both OS's http://www.rwlabs.com/article.php?id=732
Intel SSD Tool's manual TRIM works on raid0 with latest OROM and drivers on my Asus Sabertooth P67, and do indeed have an positive effect. The sad part is that it's Intel SSD's only, like with singelSSD, and likely a slim comfort for those with other SSD brands. No Z77 for me I decided long ago, but this helps in low willpower moments, when explosive sales drops into my inbox.. :P
The 11.6 OROM/drivers scored 8.5k in Anvil bench before manual TRIM, and 9.2k after. The positive thing is it maintains 9.2k every run after an manual TRIM, while unTRIMMED drops a little for each run done, naturally..
The 11.6 iaStorA/F SCSI and all others of it's kind, are a little slower then the 11.2 iaStor drivers. So for optimal benches one might revert to 11.2 after using 11.6 for TRIMMING, allthough removing the pesky leftovers from SCSI Intels, are a drag at best of times.. It scores 9.7k after reverting from 11.6 and manual TRIM vs. 9.2k with 11.6 and for daily use one would just leave the 11.6 installed as one would never notice the difference.
I might soon SE the drives back to prestine performance, as I now probably can maintain that performance better.
I also noticed the better performance of 11.2 but since I don't have a Z77 chipset, I stayed with it but will gladly update to gain R0 trim. Waiting to see how much the 64GB SuperSSpeed SLCs go for but it's soo tempting just to add another 520.
guys i need some help here
im runing on M5F raid of SSD on sata 3 intel ports and 1 3TB Seagate on the ASMEDIA sata 3 also, when i open intel RST UI gives an error that stopped working, i reinstalled deleted installed again...maybe it fails cauz it tries to read the ASMEDIA ???
It's the updated version of the ISDI2.dll, which causes the problem. You may be able to solve the problem by replacing the DLL file by a previous version. For further details yo may look into >this< discussion.
Fernando i replaced and now it doesnt start says thats an error and that i should restart or reinstall program...
wich version is best to replace ?
**i replaced with a lot and none works
@ all users with an Intel 6-series chipset RAID0 SSD system (incl. X79):
If you are waiting for the "TRIM in RAID0" feature, you should have a look into >this< very interesting thread at AnandTech Forum.
According to Dufus he managed it to get TRIM working within his P67 RAID0 array by just modifying the Intel RAID ROM module v11.6.x.xxxx!
These test results verify, that Intel should be able to create an updated RST RAID ROM version, which offers the "TRIM in RAID0" feature for much more Intel chipsets than just the 7-series ones.
Can someone plz tell him to post this modified OROM, or how to? The guy is actually afraid Intel will hit him with the "hammer" if he posts this, gee... It's not illigal to modify oroms or bioses, never have been or never will be, period...
The mainboard manufacturers ASRock and GIGABYTE have published the first X79 mainboard BIOSes with the ability to flash an Intel RST RAID ROM of the v11.6 series. A user named snout, owner of a GIGABYTE G1.Assasin X79 board, just has verified, that TRIM is working within his X79 RAID0 system after having flashed the new BIOS containing the Intel RAID ROM v11.6.0.1702. For details you may look >here<.
I posted this on the asus rog site to see if anyone knows about the RIVE and a possible update bios. Thanks Fernando for the update.
Good news also for people with 6 series MB. It has been confirmed that trim can be enabled by modding the latest Intel orom. Details the thread on Anandtech which I think was mentioned earlier.
i'd like to thanks to Fernando, Dufus/CPL0, Snout and others for the amazing work on this subject and hope this would stimulate Intel to give raid0 TRIM command for SSD users of all their platforms that support it and not only the most recent ones.
All the work that has been done here is truly appreciated, especially for people like myself, that do not have the skills you all have, but still have the passion to see our systems run to their fullest potential!
With that said, my head still hurts from all the info and miss-info from forums around the World regarding the whole RAID / TRIM support for the X79 chipset.
And due to my lack of thorough research, I unfortunately have a New Egg order on the way that includes the Sabertooth X79 MB, Intel i7 3820, and 2 OCZ Vertex 4 (256) SSDs.
So, my options as I see them:
1) Return one of the drives.
2) Keep both, using one for storage and HOPE an RSTe update is released by Intel soon.
3) Refuse the shipment and reorder an i7 3770k and ASUS P8Z77, that will allow RAID/TRIM support.
4) Other ideas.....
I'd really like to keep the board as the new Ivy Bridge-E that is supposed to be released Q3 2013 would fit nicely in the Sabertooth socket 2011, while the 1155 (ASUS P8Z77 mentioned above) socket is phased out. So, IF it is possible for someone to walk me through the process mentioned in this forum that would allow me to keep my NewEgg order as is, and take advantage of a RAID with TRIM it would be greatly appreciated. Thanks/Danke.
In th big scheme of things, most modern SSDs in R0 do fine with no TRIM. It's nice to have, but I'm not giving up my 2600K OC and spending a ton of money to get it. IB gives me nothing over my current SB setup.
Haswell will be here in April. That's when I upgrade.
Honestly, I'd wait for Haswell, but you haven't told us your current setup. What do you currently have?
Well again I failed to do the legwork, and just read about the Haswell today but the release date is "sometime" 2013, and what sparked my spending spree was Newegg offering the OCZ Vertex 4 for $159....$90 off their regular price of $249.....So, I took the leap and ordered what I posted earlier, not knowing that the MB would not support the TRIM in RAID......With that said............
My current system has the 1366 i7 980x running on an EVGA X58 chipset. The Newegg order i7 3820... based on AnandTech benchmarks is an even swap compared to the 980x, but the ASUS X79 has pcie 3.0, an SATA improvements. I know I should have done more reseacrh b4 jumping into the fire, but now that I'm here I'm trying to get the best out of what I have and have the options stated in my previous post, with one more that I failed to think about:
5) Send one of the SSDs and the i7 3820 back, and get the i7 3930k......
Well, now you can have the cake and eat it to, I've just confirmed TRIM-raid0 on P67 with an orom modded by Fernando: http://forums.anandtech.com/showpost...&postcount=102
If you need it, i'll replace the orom on the bios of your choice, no problem at all..
That is just EPIC!
Now check the x58 for trim raid-0 too :D
Thank you! Nice job :up:
Here is your bios: http://www.diskusjon.no/index.php?ap...tach_id=509420
Great news Ourasi! TRIM IS WORKING on my P8P67 Deluxe Vertex 3 240 RAID0 array. I have verified 5 or 6 times using the hex editor. No reboots, no recreating the array. Just instaneous. I tried 3 times before flashing your BIOS and a definite no-go - hex showed stuff was still there. I waited 2-3 minutes and it didn't happen. I flashed your BIOS, re-setup all of my settings, tested again and voila - the hex editor showed an immediate TRIM job - like 5-6 times, each time successfully and immediately.
Awesome work! I owe ya one!
Sweet, no longer degraded arrays for you either, nice... I will rate the bios I made for you as "Tested and TRIM-working", so others with the same board knows it's working.. SF drives are nice to test TRIM+HxD with, since the TRIM is immediate, while some others like C300 (tested myself) only marks it for deletion/TRIM and then TRIM it at shutdown/reboot and possibly logoff/on, quickformat-TRIM on the otherhand is immediate.
FWIW, I have the following mobos and each has the same output from the SATAID tool:
P8P67
P8P67 Deluxe
P8P67 M
Controller ID: 2822
Alternate ID: 1C04
Hey guys, I have a rampage 3 extreme and I modded the latest 1601 bios with the 11.6.0.1702 Intel sata/raid firmware.
I can confirm that raid 0 trim is working on x58.
I had to use the latest RST 12.0.0.1013 beta driver from station-drivers because I couldn't get the rapid storage utility to load on any of the others, but I would guess that any driver that supports trim would trim the drive.
http://www.station-drivers.com/page/intel%20raid.htm
Since the Intel RAID ROM v11.6.0.1702 natively doesn't support TRIM in RAID0 for any other than 7-series and C600-series (X79) chipsets, I would like to know how you have modified the Intel ROM module and how you have verified, that TRIM is actve within your RAID array.
3A16 0000 07
I figured out how to do this by following the information in:
http://forums.anandtech.com/showthread.php?t=2281849
Edit: Looks like you're one of the main posters in that thread. Hadn't noticed till now. But yeah the LPC ID for my controller is not the 3A25 as reported in that thread by other x58 users.
Also figured I'd point out that my controller does NOT show up in device manager as "82801JIR (ICH10R) SATA RAID Controller"
Mine shows up as "Intel(R) Desktop/Workstation/Server Express Chipset SATA RAID Controller"
Thanks for this very useful information.
Can you post some screenshots, which verify, that TRIM really works within your RAID0 after having flashed the BIOS with the modified ROM module?
That may be useful for other users with an X58 RAID0 system, who would like to get TRIM support too.
Hi zoson
I also use a ASUS Rampage 3 Extreme and the latest 1601 bios with the Intel OROM 11.6.0.1702.
We would appreciate it if you could verify for us, post proof that RAID0 TRIM is definitely working
on our board since doing the Intel OROM module modification. Could you also please post a
screenshot showing the hex code modification you did on the Intel OROM module for our board.
I have two SSDs in RAID0 on this board.
Thank You.
Verification:
http://i144.photobucket.com/albums/r...d-c-before.jpg
http://i144.photobucket.com/albums/r...ed-c-after.jpg
I just make a text document at C with unique text and search. Then delete the txt file.
I think it's easier than using a jpg and less chance of duplicate text.
zoson, Any chance you could please post a screenshot showing the actual hex code modification that you did
on the Intel OROM module for our board just like others have done in the thread on AnandTech so that us that
are not really familiar with using a hex editor knows exactly what FLAGS (hex code pairs) to change to what?
Or if you could share with us your modified Intel OROM 11.6.0.1702, that would be great.
Also what OS are you using to do your tests?
Thanks a lot.
zoson, have I got this modification correct, exactly 100% the same modifications you made?
Screenshot:
http://imgly.net/img/jHkzc.png
This is my first time trying this and so don't know whether i've done it correctly or not.
By the way, I take it your test you posted above always failed, TRIM not working before you
did the hex code modification to the Intel OROM module?
@ Sylver123:
Before you are trying to flash any BIOS with a modified Intel RAID ROM module, you should read >this< statement from Dufus at AnandTech Forum regarding the TRIM test done by zosan.
I have to agree with you Fernando, and Dufus, that looks not TRIMMED to me either.
I'll hand out some advice when it comes to using .txt files for TRIM-testing. This is an example on what to write in the txt file:
http://bildr.no/thumb/1328585.jpeg
Then it will look like this before delete/TRIM in HxD:
http://bildr.no/thumb/1327527.jpeg
And after delete/TRIM when it's working:
http://bildr.no/thumb/1327528.jpeg
Always search for the text in the middle of the .txt file, in my case "Betty er snill", as then you'll find the blocks and offset in the middle of the file, and after TRIM the whole viewable page is cleaned out in HxD.. Fill an A4 page atleast, so the file is a decent size...
Happy Trimming fellas... :P