The whole point of professional data recovery tools is to make the recovery process faster and more simple. USB stabilizer does just that. And, with its pass through ability for VSC, we can work directly with firmware on drives that aren't locked down.

_________________
Luke
Recovery Force Donor Inventory: 5,000+ Fully Cataloged HDDs & Growing

That suggests that the stabilizer transparently resets the drive when it it hangs up, and Windows is unaware that this is happening. I can see why you would need such a tool if there were no way to intercept the OS, but HDDSuperClone seems to be able to do this just fine under Linux. It would be interesting to benchmark this device against HDDSuperClone in a Linux environment.

_________________
A backup a day keeps DR away.

[Unlocked] USB drives all support pass through for VSC, so the stabilizer adds nothing in this regard.

_________________
A backup a day keeps DR away.

Indeed, that's the idea as far as I can tell. There's a number of options you can configure, all enabling you to control what Windows is allowed to see and handle. Resets and such are handled by the DeepSpar and you can select if you want Windows to know about these. You can even make Windows believe writes were successful even though they're intercepted and never reach the drive.

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

So, HDDSuperClone will just pause and wait for the drive to redetect if it is physically disconnected and reconnected to the system?

_________________
Luke
Recovery Force Donor Inventory: 5,000+ Fully Cataloged HDDs & Growing

Yes, but what Luke tries to tell, I think, is that it allows Stabilizer to work 'in tandem' with PC3000, MRT and such. Well, with pretty much any Windows tools, as I said earlier.

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

So basically Windows is too stupid to handle errors gracefully, and the stabilizer keeps it out of the picture by lying to it? This begs the question, why do Ace and others choose Windows as their platform?

In any case, there was a discussion in another forum which turned up an interesting titbit, namely that Windows uses a single timeout value for all disk I/O (in the registry), including spin up. I wonder what would happen if this timeout were reduced to its minimum value, ie 1 second. The default is 60 seconds (= 0x3C). Other people referred to an error recovery timeout of 600ms or 500ms, but no-one was able to back it up with references.

_________________
A backup a day keeps DR away.

Serge at DeepSpar asked if I could post this for him. He is having issues with his hddguru account.

You don’t seem to understand how the tool works, I guess you did not watch the demo video? This is the most complicated architecture we have ever built and we had no choice but to build it because the core functionality is not even theoretically possible to implement with software alone. Look at what we show at 3:13-3:31 in the demo video: https://youtu.be/TqxMUtF1vPY?t=193. That is shown specifically to prove that it is a true hardware solution – it is completely impossible to do this with software alone... Windows/software is connecting to the ARM processor onboard our hardware which is emulating the actual source drive. There is never direct access from the software tool to the drive and this kind of hardware separation is the only way to properly do read instability handling, i.e. the only way to guarantee that temporary source drive issues do not cause software/Windows to drop the drive and abort the process (or hang, blue screen, etc.). The ARM processor works to reliably maintain the connection regardless of source drive status, protecting the rather fragile Windows environment from the drive.

Separating the drive on the hardware level is also the only way to do hardware USB resets and repowers without aborting the recovery process because doing those things would of course close the USB connection every time. But even before that point, there is no way to implement reliable hardware USB resets with software alone, so all the speed-ups that users are showing in this thread would not be possible. USB resets are not even a little similar to SATA resets… From the last decade of doing research on this subject I can confidently tell you that the only reliable way to reset USB devices is to do a PHY reset, i.e. physically disconnect the data USB pins for a number of milliseconds then reconnect them, which can also only be done with hardware. You can quite literally do none of what we are doing with this product with a software-only solution, even in theory…

Being able to focus on specific files allows recovering drives that re in much worse shape, in comparison to being forced to do a full clone. Being able to use the product with any software tool opens up more possibilities for file recovery, which directly improves recovery rate. Being able to work with vendor specific commands is also a nice bonus because it allows tools like PC-3000/MRT/DFL/etc. to be more successful at applying firmware solutions (like WD Slow Responding fix) over USB. Of course in the majority of cases when such a fix fails over USB it is because the communication broke down inside the drive between the SATA HDD and USB bridge, but in a notable minority of cases communication breaks down between the USB bridge and Windows, which our device would help with, making USB fixes more likely to succeed.

_________________
Luke
Recovery Force Donor Inventory: 5,000+ Fully Cataloged HDDs & Growing

I am a happy user of DeepSpar USB Stabilizer and to me it offers added value. I have no further interest in 'defending' the tool, other than telling you it helps me. I think HDDSuperclone is a cool tool, but not for me.



Yes.

Or you could argue that it's not too far fetched for an OS to expect healthy hardware. And an intact file system. I mean, where do you want to draw the line? It's a design choice I guess, to wait x time on error, to try to mount a file system by default etc.. Perhaps Windows could handle bad sectors in meta data (I think this is what is the core cause of freezing) more gracefully and rather than lock up simply quit trying and pop up a message telling there's a problem with a device which prevents Windows from working with it. But even then USB Stabilizer would have value.



I am not so sure this covers all disk IO. This says it affects SCSI Miniport driver: https://docs.microsoft.com/en-us/window ... rt-drivers. TBH I don't know what would be affected by that.



I am convinced I read this 600 ms value somewhere too, but I must admit I haven't been able to find where I read (or heard, I think it was a video) this.

But even if so, 1 second or 5 seconds is still very coarse and long. And then there's retires Windows may do which can also be intercepted by the Stabilizer. Is one second a reasonable time out on a severely degraded drive? Or for a USB flash drive? And again, even you could limit to 1 second, it still will do retries. USB Stabilizer allows you to prevent those. And do resets, power cycles.



You'd have to ask them. Maybe it was better or less of an issue with previous versions, I have no idea. Maybe because many projects started on an attic or in a garage and it happened to be developed for Windows, by accident. And maybe not all things aren't decided on all that rationally. Who knows. Even relatively young products are developed for Windows while they could have opted for Linux. Would VNR be a better VNR and more successful VNR if it ran in Linux?

Anyway, since most tools run in Windows it makes perfect sense DeepSpar designs a tool to handle unstable drives in that Windows environment.

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

I would just like to point out that the direct USB mode of the pro version of HDDSuperClone does not yet support SD cards. That requires implementing a special SD to SCSI protocol (why did they have to do that?!), which is on my bucket list to do. Most if not all other USB attached drive types should work, and I would be very curious to see it compares to the USB Stabilizer in those situations.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

I confess that the stabilizer is a lot more than just a dongle. I guess I was playing my part as the Devil's Advocate a little too enthusiastically. :-)

That said, I still take issue with the claims regarding the transport of VSCs over USB. They seem to be implying that the stabilizer adds this functionality, which is misleading. In fact it would be more appropriate to say that the stabilizer doesn't interfere with an existing feature. After all, sending ATA commands, whether standard ATA or VSC, to PATA/SATA HDDs over USB via "pass through" has been possible for over a decade. In fact HDDSuperTool does this quite well. Furthermore, all SMART tools use this standard feature to retrieve SMART data with standard ATA commands.

HDDSuperClone can control the power to a storage device via a relay, so that's not a unique feature. That said, the OS would be aware that the drive disappeared whereas this would not be the case with the stabilizer. I suppose one could implement a PHY disconnect in similar fashion or with an IC switch, but with the same caveat. In fact MRT has such a switch, although it comes preconfigured as a placebo. :-)

_________________
A backup a day keeps DR away.

Power control helps a lot and no, it's not unique. This is my own clumsy way of handling it https://youtu.be/v1ql6yNS0qk?t=89

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

That's brilliant!

https://www.yepkit.com/product/300115/YKUSHXS



Now for a USB 3.0 version ...

_________________
A backup a day keeps DR away.

They have a USB 3 version too if I'm not mistaking, I just didn't pay attention when I purchased.

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

You're right:

https://www.yepkit.com/product/300110/YKUSH3

_________________
A backup a day keeps DR away.

Before I post too much more, I want to be clear, I do not have a USB Stabilizer. But I can see that it is a very valuable device, and if someone had the money, it could/would be a good investment. It is a great, if not excellent tool. Now that I have that out of the way...

When it comes to power cycles (and resets), yes, that is great. But when the drive is having issues, it can lock up even the best OS. Without the ability to set a timeout, you are stuck with what the OS does. The USB Stabilizer handles the timeouts internally, so the OS does not even see it. That is why it would be superior over an external relay system using the OS for reading. That means the USB Stabilizer can do amazing things when trying to read an unstable device through the OS. That is why it is great.

The direct mode of HDDSuperClone takes it a bit farther, and removes the OS from the equation. Once in the direct USB mode, the OS does not see the drive, which keeps it from trying to read it, and allows HDDSuperClone the opportunity to clone/image it (or access it through the virtual driver mode). It handles the timeouts, and allows for using an external relay to power cycle the drive without the OS being involved. But at this time it can’t handle SD card readers (that is on the to do bucket list).

And wow, that Yepkit USB Switchable Hub looks great, and it works with Linux! I might have to buy one of those to play with. I will put that on the bucket list.

For the most part, the USB Stabilizer can be better than HDDSuperClone, with the simple fact that it works with all attached devices (as far as I know), and works with Windows, which is what everyone uses. But there is one case where I do know the Direct USB mode of HDDSuperClone would be superior, and that is working with a slow responding WD USB drive without converting to SATA. It allows the ability to increase the cluster size to a point that greatly speeds up the cloning. Try that with the USB Stabilizer

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

If the stabilizer is a hardware tool, then why does it "Require Windows 10 x64"? Why not Linux, or Mac OS, or any other OS?

The web site offers "adapters for SATA, NVMe PCIe M.2 M-key, AHCI PCIe M.2 M-key, and Apple 12+16 pin PCIe SSDs". How do these adapters differ from the "basic adapters"? Do Deepspar's adapters allow for disabling bad PCIe lanes, changing link speeds, and issuing resets whereas the basic adapters do not? Or can the stabilizer achieve the same control with basic adapters?

http://www.deepspar.com/blog/Recovering-Problematic-PCIe-SSDs.html

_________________
A backup a day keeps DR away.

But Stabilizer is not a cloner. So you'd need a Windows based imaging program that'd allow you to experiment with that. Or isn´t that what you mean? How does this work? You gradually increase size and then slightly lower until optimum speed is reached?

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

Yes. Stabilizer doesn't do all that AFAIU (disabling bad PCIe lanes etc. - I think this refers to DDI PCIe SSD Add-on)

_________________
Joep - http://www.disktuna.com - video & photo repair & recovery service

The web site seems to imply that there is something special about the "set of adapters" that Deepspar offers as an optional extra.

http://www.deepspar.com/usb-stabilizer.html


The implication seems to be that Deepspar's adapter are "not basic". Do users really need to purchase these adapters to get maximum benefit out of this tool?

_________________
A backup a day keeps DR away.